News from Industry

The Hidden AV1 Gift in Google Meet

webrtchacks - Tue, 12/19/2023 - 15:06

Earlier last week a friend at Google reached out to me asking Does Meet do anything weird with scalabilityMode? Apparently, I am the go-to when it comes to Google Meet behaving weirdly :). Well, I do have a decade of history observing Meet’s implementation, so this makes some sense! It turned out that this was […]

The post The Hidden AV1 Gift in Google Meet appeared first on webrtcHacks.

Twilio exits video APIs, further focusing on voice, SMS and Segment

bloggeek - Wed, 12/06/2023 - 09:35

Twilio Programmable Video is no more. What should WebRTC Video API vendors and their customers do from here on?

This week, Twilio dropped a bombshell

It decided to shut down its Programmable Video service and do a bit of downsizing and trimming around Segment and Flex.

I didn’t intend to write anything more until 2024, but this necessitated changing my plans.

The image above is an adaptation from a blog post on Twilio’s website from 2021…

Table of contents

Twilio Signal, and why I stopped covering it
CPaaS vendors: Best of breed vs best of suite
The cases of Twilio IOT and Twilio Live
The demise of Twilio Programmable Video
Innovations in Video APIs and WebRTC managed services
The rise of the Zoom Video SDK
The future of managed Video APIs (without Twilio)
Where should Twilio Video customers go from here?

Twilio Signal, and why I stopped covering it

Each year, Twilio hosts its Twilio Signal event. I’ve attended a couple of them in person and used to cover them here on a yearly basis.

That stopped with Twilio Signal 2021, which was the last time I covered that event here. The reason for that was the pivot Twilio made from CPaaS to CEP (Customer Engagement Platform).

Ever since, I’ve searched for things to talk about and share about Twilio Signal, but found nothing of real value or interest to my readers.

Remember – I cover WebRTC and CPaaS. CPaaS mainly from the point of view of WebRTC and modern communications and less from the SMS and legacy telephony sides of it.

The shift towards CEP meant a lot less investment and focus by Twilio on exactly these areas – WebRTC and CPaaS that are non-SMS/legacy telephony related.

What did Twilio have to show for its investment in video and WebRTC in 2022 and 2023? Nothing. Crickets. Oh… yes… they did integrate with Krisp for noise cancellation. Presumably only in their Video SDK and not the Voice SDK. So that’s down the drain as well.

The decision might be the right one for Twilio, if you look at where their investments and attention are going:

Twilio Flex, for a programmable contact center
Segment, as a leading CDP vendor
Fuzing Segment with programmable communications

Video is likely 1% or less of their revenue. So why bother? Especially when it requires management attention to get it anywhere meaningful with so much else that is bigger and more important to deal with.

CPaaS vendors: Best of breed vs best of suite

I learned about the concepts of best of breed and best of suite when working at Amdocs.

A best of breed vendor would specialize vertically, offering its customers a solution that is great in a narrow domain. Think of it as “the leading SMS vendor”. You do SMS and only SMS and you do it really well
Best of suite is all about the breadth of your offering. You provide a solution that has a mixture of multiple services and features your customers will need. You might not be doing any of them the best in the market, but if someone needs multiple services and wants a single vendor to work with – you’re the best for them. Think of it as offering SMS, voice, email, video, … – Twilio

Twilio started with SMS and voice. It later decided to expand and become “best of suite” by attaching to it email, video, IOT, social messaging, chat , …

What happened though is that in parallel, it worked hard on being best of breed in voice and SMS. Doing that by going upstream and introducing Flex. Flex reduced the effort of contact centers built on top of Twilio.

And then they pivoted. With the acquisition of Segment and the need to tightly integrate it with their CPaaS and Flex offering. Transitioning from taking care of communications to taking care of understanding the customer.

Today?

There are two types of CPaaS vendors:

The best of suite ones, who offer the breadth of communication services
Or the best of breed ones, who focus on a specific domain. And the domain I care about is WebRTC and video. These usually won’t have legacy telephony. At most, they will enable connecting to legacy telephony of third parties

Interestingly, both are circling like vultures around Twilio to see which customers are going to come out of there looking for alternatives. Some of these CPaaS vultures offer pure WebRTC video solutions. Others offer the whole suite. And there are those who don’t even offer video – but see this as an opportunity to poach customers from Twilio.

The cases of Twilio IOT and Twilio Live

I remember that in one of the first Twilio Signal events, Jeff Lawson stood on stage and proudly announced that they never deprecated an official API. The way this was later handled is by having beta and GA phases for products.

This cannot be said anymore… by the end of 2022, Twilio started sunsetting and shutting down services.

It started with a round of layoffs at Twilio. Jeff Lawson, Twilio’s CEO, wrote a message that got to the Twilio blog as well. Here’s what we shared about it at the time with our WebRTC Insights clients:

Twilio laid off 11% of their workforce
The decision was to take the internal email and publicly put that on their blog, instead of getting it indirectly on TechCrunch
A few interesting to note in this email:
- Twilio has 4 focus areas: reliability+trust, profitability of messaging, Segment adoption, Flex customer base
- 3 main products in focus: messaging, Segment (Customer Data Platform), Flex (Programmable Contact Center)
- Programmable Video isn’t prioritized at all. Programmable Voice might be said to be buried somewhere in there under Flex
- Twilio’s future success and growth lies Segment and Flex – not in Communication APIs
The charts below show the number of employees and growth rate of Twilio in recent years

Why is Twilio doing this? A few options here
- Growth is slowing, and all the hiring they did is just too much to maintain
- Management has too many directions it is now looking at, so it was time to shoot down all the smaller initiatives and products since they won’t bring the necessary growth at Twilio’s size
- Twilio might have used the current market state to clean the stables and remove all the useless fat from the company
- All of the above, to some extent
How will this affect other CPaaS vendors? This is hard to say. Here are a few thoughts
- If Twilio is in poor shape, then the rest are in worst one
- With Twilio management shifting focus elsewhere, the API space, and especially in voice and video, it is down on these areas to build some differentiation
- Time to use FUD in the market against using Twilio for video APIs – Jeff just said it isn’t a focus area. Just make sure it doesn’t backfires…
- Maybe CPaaS isn’t as great as it was believed to be as a business…
  - From my past life I know that selling to developers is super hard
  - And the target market for it is rather limited
  - There are better opportunities out there, which is why many CPaaS vendors are following in Twilio’s steps when it comes to Flex
Also, if you are looking for developers, it might be worthwhile to try and poach a few of those who still work at Twilio, or more easily those who are looking for a new job

After the reduction in workforce, came the reduction in product offerings. The first two to go through the chopping block were Twilio IOT and Twilio Live.

Twilio Live was announced dead in November 2022. Low traction of the service and little fit the the direction of Twilio meant this had to die. The way this was done? Let customers know. Officially suggest they go use Mux instead. Somehow, the fact that Mux at the time had a service competing directly with Twilio Programmable Video wasn’t something that worried Twilio.

Twilio IOT was simply sold off to KORE Wireless in March 2023.

Remember that suggestion we gave about FUD in the market against using Twilio for video APIs? (I marked it in yellow above so you won’t miss it)

The demise of Twilio Programmable Video

Here’s what the Twilio product menu looks like on their homepage:

This is likely going to change soon or by the time this gets published.

Customer Data = Segment offering
Communications = CPaaS
Applications = Enterprise stuff

Each and every piece in the Communications part can be snuggly fit into the products on the left and on the right (Customer Data and Applications).

Video is a bit of a stretch. At least if you look closely at traffic sizes and revenue numbers.

The two other oddballs – IOT and video streaming – were thrown out without too many objections and without hurting Twilio’ bottom line.

What was left was to get rid of the video piece. It likely took too many resources but made no real dent in Twilio’s numbers.

To be frank – the problems likely started with the acquisition of Kurento. Kurento wasn’t fit for what they had in mind for it, and it was riddled with architectural and technical issues. This wasn’t a good starting point for multiparty calling in Twilio Programmable Video.

If I had to guess, a lot of technical debt went into the product to improve and repurpose the media server pieces of Kurento.

Twilio was slow to innovate on video, leaving the room for other vendors – big and small. It missed the lowcode and embeddable experiences that are now common in video APIs. They didn’t invest in AI integrations too much. It didn’t optimize media quality enough to work well for its customers.

And then it left the door open for Amazon with their Chime SDK to threaten them in this domain.

I am guessing growth and revenue from Twilio Programmable Video wasn’t in line of expectations (unsurprisingly). The current market climate, the end of the pandemic, the headaches in Segment and Flex. All of it got them to the conclusion that it would be simpler to just sunset Twilio Programmable Video and move on.

A brave decision. Twilio Programmable Video couldn’t have been sunset in the worst time (unless you consider a few months prior to the pandemic and the quarantines).

A week before this announcement from Twilio, Amazon announced support for video calling in Amazon Connect.

Amazon is investing in adding video to its contact center solution, and Twilio, who has Twilio Flex competing against Amazon Connect, is sunsetting video support for its video API.

What does it mean for video calling support in Twilio Flex?
Would Twilio still support or add video calling to Twilio Flex without offering Programmable Video APIs?
How should contact center customers view this? If they have video requirements in their roadmap, would they use Amazon Connect or Twilio Flex?

Innovations in Video APIs and WebRTC managed services

Why was Twilio Programmable Video appealing to potential customers? I can think of two main reasons:

Single throat to choke. Sourcing your voice, SMS and video from the same vendor, on a single bill is an advantage
A reputable vendor. It is Twilio. They are big. What can ever go wrong? …

The reasons why not to? Quite a few:

Quality wasn’t on par with what can be achieved elsewhere with CPaaS vendors
No lowcode/embeddable offering for its video API
Support… could be better
No innovation

All that Twilio had for itself is its brand name. And that in a market that was moving on.

Things other vendors have been doing in that period of time?

Doubling down on large scale sessions, with 10,000 or more users
Live streaming solutions (the one Twilio sunset in 2022 – Twilio Live)
Investing in AI integrations and pipelines, both on client side and on server side
3D audio, VP9 video codec support
Nocode/lowcode solutions

Twilio wasn’t able to keep up. Or even pick a direction it wanted to invest in.

The rise of the Zoom Video SDK

Twilio issued an email to its customers on December 5, stating the sunset will take a full year. From this email:

[…] we have decided to End of Life (EOL) our Programmable Video product on December 5, 2024, and we are recommending our customers migrate to the Zoom Video SDK for your video needs.

The official recommendation from Twilio is for their customers to migrate to the Zoom Video SDK.

The announcement can’t be found (yet) on any marketing material from Twilio. It can be found on social media accounts from Zoom.

Why Zoom?

Zoom isn’t a competitor of Twilio in anything, and are unlikely to be any time soon
It is a large and respectable vendor with a brand name

They couldn’t suggest vendors that have SMS or voice services.

The rest are mostly smaller vendors – not something Twilio wanted to be identified with is my guess.

There’s only one problem with picking Zoom Video SDK here. Their web experience isn’t on par with the rest of the pack. They rely on WebTransport+WebCodecs+WebAssembly, which isn’t as stable or performant as just using WebRTC. For native, their SDKs should be fine, but for web browsers, I’d be reluctant to use them yet. Add to that the fact that this is a technology shift, requiring some relearning of terms and a reliance on proprietary technology, and you get some increased risk for the vendors switching.

I wonder if Twilio and Zoom came to an agreement here (with Zoom maybe even paying for this suggestion to go out) or if Twilio simply decided to offer some kind of a recommendation and be done with it. Philipp’s bet: Eric had dinner with Jeff and paid for it.

Anyhow, customers have a full year to figure out a solution. Or less – depending on how much browsers WebRTC implementations drift away from the current implementation of Twilio. What doesn’t get maintained in WebRTC rots rather quickly.

The future of managed Video APIs (without Twilio)

I am not sure how much Twilio Programmable Video would be missed.

Developers certainly used it. Big and small. Its revenue was probably higher than some of the smaller video API vendors out there. These developers will figure out a way to migrate to other vendors to use. It won’t be the first time a CPaaS vendor has existed in the video API market (we had AddLive, vLine, ooVoo, SightCall, Respoke, Tropo, Forge, CafeX, Circuit, Bit6 all exit this market in the past).

3-4 years ago, we had 3 top dogs in this market: Vonage, Twilio, Agora

A year ago, I’d say I heard a lot more about Vonage, Amazon Chime SDK and Twilio. Less so Agora

Now, we have Vonage and Amazon Chime SDK

Who will take the 3rd spot in the 3 runners when it comes to developers’ mindshare in this industry?

We have Agora, Daily, Dolby, LiveKit and others who are all vying for that spot. Each has its own angle and differentiation.

Would Vonage keep its spot there?

Will Amazon continue investing in its Chime SDK enough?

I don’t have the answers to these questions, but I do have my own opinions.

Where should Twilio Video customers go from here?

That is the big question.

If you are using Twilio Programmable Video – who should you go to instead?

And if you are on the lookout for a CPaaS vendor now – who should you pick?

My WebRTC Developer Landscape infographic was last updated in 2022, but can still offer some guidance as to the alternatives available. Some of them I’ve listed throughout this article. Others are just as valid.

Here are a few questions you need to answer for yourself:

What are your requirements and focus? Different CPaaS vendors offer a different type of a solution, so pick one that offers what it is you’re after
Make sure you ask around. Check references. Talk with other developers who use that CPaaS vendor
Try them out in a small POC before fully committing yourself
Check their commitment and level of investment in what it is you focus on as your requirements and roadmap. Don’t only listen to what they say – also check out what features they introduced to the market in the last 12-24 months. See if they had layoffs in that same period of time as well
Don’t invest in abstraction layers to be able to replace CPaaS vendors. It sounds like a great initiative and project, so just don’t do it. Unless you want to use more than a single vendor at a time (unlikely for most of us)
While you shouldn’t invest in an abstraction layer, you should definitely try to limit calls to the CPaaS vendor’s APIs to specific modules in your code. If you can limit it to a single source file or class – even better

The post Twilio exits video APIs, further focusing on voice, SMS and Segment appeared first on BlogGeek.me.

Third time’s a charm: WebRTC Insights, 3 years in

bloggeek - Mon, 11/20/2023 - 12:30

Let’s look at what we’ve achieved with WebRTC Insights in the past three years and where we are headed with it.

Along with Philipp Hancke, I’ve been running multiple projects. WebRTC Insights is one of the main ones.

Three years ago, we decided to start a service – WebRTC Insights – where we send out an email every two weeks about everything and anything that WebRTC developers need to be aware of. This includes bug reports, upcoming features, Chrome experiments, security issues and market trends.

All of this with the intent of empowering you and letting you focus on what is really important – your application. We take care of giving you the information you need quicker and in a form that is already processed.

Three years into this initiative, this is still going strong. We’ve onboarded a new client recently, and this is what he had to share with us on the first week already:

“[The Insights] Newsletter has been great and very helpful. Wish we had subscribed 2 years ago.”

Sean MacIsaac, Founder and EVP, Engineering @ Roam

Why is the WebRTC Insights so useful for our clients?

It boils down to two main things:

Time
Focus

We reduce the time it takes for engineers and product people to figure out issues they face and trends on the market. Instead of them searching the internet to sift through hints or trying to catch threads of information on things they care about, we give it straight to them – usually a few days before their clients (or management) complains about it.

On top of it, we increase their focus on what’s important to them. Going back to past issues to find problems, search issues, look at security problems, know of experiments Google is doing or just be aware of the areas where Google is investing their efforts – all of these become really simple to do.

In the past few weeks we’ve been getting complaints from clients about audio issues on Mac (usually acoustic echo problems in Chrome). These were already hinted to in one of our previous issues and the full details appeared in the more recent issues. In parallel, we’ve been able to sniff around for root causes for them almost in real-time – enabling them to zero in on the problem and find a suitable workaround.

If I weren’t so modest, I would say that for those who are serious about WebRTC, we are a force multiplier in their WebRTC expertise.

WebRTC Insights by the numbers

Since this is the third year, you can also check out our past “year in review” posts:

This is what we’ve done in these 3 years:

26 Insights issued this year with 329 issues & bugs, 136 PSAs, 15 security vulnerabilities, 230 market insights all totaling 231 pages. That’s quite a few useful insights to digest and act upon.

We have covered over a thousand issues and written more than 650 pages.

WebRTC is still ever changing – both in the codebase and how it gets used by the market.

Activity on libWebRTC has cooled down yet again in the last year, dropping below 200 commits a month consistently:

This is more visible by looking at the last four years:

On one hand WebRTC is very mature now, on the other hand it seems to us that there is still a lot of work to be done and bugs to be fixed. External contributions were up. What is concerning is that the “big drop” in May happened three months after Google announced a round of layoffs but we have not seen many departures of long-time contributors.

Let’s dive into the categories, along with a few new initiatives we’ve taken this year as part of our WebRTC Insights service.

Bugs

The number of reported external bugs has dropped considerably as did the number of issues tracking new work and initiatives. This correlates with the decreased commit activity.

The areas for bugs also shifted, we have seen a lot more issues related to hardware acceleration (since Google is eying that now to further reduce the CPU usage in Google Meet). Operating systems are starting to become a bigger issue, for example MacOS Sonoma caused quite a few audio issues and enabled overlaid emoji reactions (a bad choice with consequences described here) by default as part of a bigger push to move features like background blur to the OS layer. And of course, every autumn brings a new Safari on iOS release which means a ton of regressions…

A good example of how Philipp himself uses Insights as a way to identify what change caused a regression was the lack of H.264 fallback on Android which rolled out in Chrome 115 in August. We had been commenting on the original change end of May:

That said, we did not think of Android which remains complicated when it comes to H.264 support. Thankfully this rollout was guarded by a feature flag so the regression could be mitigated by the WebRTC team in less than two days.

PSAs & resources worth reading

In addition to the public service announcements done by Googlers (and Philipp) as part of making changes to the C++ API or network behavior we continue to be tracking Chromium-related “Intents” (which are a useful indicator for what is going to ship) and relevant W3C/IETF discussions in this section. We also moved more in-depth technical comments on relevant blog posts from the “Market” section which made the overall decline in activity less visible here.

Experiments in WebRTC

Chrome’s field trials for WebRTC are a good indicator of what large changes are rolling out which either carry some risk of subtle breaks or need A/B experimentation. Sometimes, those trials may explain behavior that only reproduces on some machines but not on others. We track the information from the chrome://version page over time which gives us a pretty good picture on what is going on:

We have gotten a bit better and now track rollout percentages. We have not seen regressions from these rollouts in the last year which is good news.

WebRTC security alerts

This year we continued keeping track of WebRTC related CVEs in Chrome (15 new ones in the past year). For each one, we determine whether they only affect Chromium or when they affect native WebRTC and need to be cherry-picked to your own fork of libwebrtc when you use it that way.

In recent months we’ve seen a trend of looking more closely at the codec implementations to find security threats there. Our expectation is that this will continue in the coming year as well – expect more CVEs around this area.

A personal highlight was Google’s Natalie Silvanovich following up on a silly SDP munging thing Philipp did with CVE-2023-4076 which affected WebRTC munging in Chrome (but not native applications:

If only anyone had told us that using SDP in the API, let alone having Javascript manipulate it in the input, is a bad idea…

WebRTC market guidance

What are the leaders in video conferencing doing? What is Google doing with Meet, which directly affects WebRTC’s implementation? Are they all headed in the same direction? Do they invest in different technologies and domains?

How about CPaaS vendors? How are they trying to differentiate from each other?

Other vendors who use WebRTC or delve into the communication space – where do they innovate?

Here’s a quick example we’ve noticed when Twilio worked on migrating their media servers to different IP and ports:

This ability to look at best practices of vendors, how they handled such challenges, or introduced new features is an eye opener. These are the things we cover in our market guidance. The intent here is to get you out of your echochamber that is your own company, and see the bigger world. We do that in small doses, so that it won’t defocus you. But we do it so you can take into account these trends and changes that are shaping our industry.

The interesting thing is that as WebRTC goes more and more into a kind of a “maintenance mode” with its browser releases, the variance and interesting newsworthy items we see on the market as a whole is growing. This is likely why our market insights section has seen rapid growth this year.

Insights automation

We’ve grown nicely in our client base, and up until recently, we sent the emails… manually.

It became a time consuming activity to say the least, and one that was also prone to errors. So we finally automated it.

The WebRTC Issue emails are now automated. They include the specific issue along with the latest collection security issues. It has made life considerably simpler on our end.

Join the WebRTC experts

We are now headed into our fourth year of WebRTC Insights.

Our number of subscribers is growing. If you’ve got to this point, then the only question to ask is why aren’t you already subscribed to the WebRTC Insights if WebRTC interests you so much?

You can read more about the available plans for WebRTC Insights and if you have any questions – just contact Tsahi.

Oh – and you shouldn’t take only our word for how great WebRTC Insights – just see what Google’s own Serge Lachapelle has to say about it:

Still not sure? Want to sample an issue? Just reach out to me.

The post Third time’s a charm: WebRTC Insights, 3 years in appeared first on BlogGeek.me.

Qotom Q20321G9 fanless PC

TXLAB - Tue, 11/07/2023 - 00:04

As PCengines announced the end of sales of their famous APU platform, it’s time to look for alternative devices that can be utilized as firewalls or network probes or VPN appliances.

I bought recently a Qotom Q20321G9 mini-PC from AliExpress. The model is similar to their Q20331G9 model described on Qotom website. The difference is a slower CPU and less SFP+ interfaces:

ModelQ20321G9Q20331G9CPUIntel Atom C3558RIntel Atom C3758RTDP17W26WNICs2x SFP+, 2x SFP, 5x 2.5Gbit LAN4x SFP+, 5x 2.5Gbit LAN

Comparing to the APU platform, this Qotom box is huge: 62mm high, compared to 30mm of APU enclosure, 217mm bright, and much heavier because of the massive heatsink. But it has much more to offer.

Two M.2 NVME sockets allow a redundant storage setup out of the box. Also, it supports ECC RAM (although the model I received had a non-ECC DIMM), so it can serve as a reliable hardware platform if you need a long-term service. Also, it has an M.2 socket for an LTE modem, two antenna mounting holes, and a nano-SIM card slot.

A minor downside is that even at idling, with all CPU cores running at 800MHz, the device is getting quite warm. The onboard sensors show the CPU core temperatures at around +42C to +44C, and the enclosure is rather hot at the touch.

I also have run a CPU stress test with the enclosure covered by a towel for about a half an hour, and the CPU temperature exceeded 60C, still functioning well.

A minor inconvenience is that the power button is too easy to press if you’re moving around it while testing. But the button is easy to remove, so that the power switch can be pressed by a pen when needed.

The SFP and SFP+ interfaces were recognized by Debian 12 out of the box.

The device arrived with a preinstalled Windows 10. The BIOS allows redirecting the console to the COM port, which is provided as an RJ-45 socket, with the same pinout as Cisco routers.

The NIC numbering is a bit non-intuitive, and the marking on the enclosure does not help much. Here are the interfaces as they’re seen by Debian, if you look at the device’s interface panel:

eno1 (SFP+)eno3 (SFP)enp7s0 (LAN)enp6s0 (LAN)enp8s0 (LAN)eno2 (SFP+)eno4 (SFP)enp5s0 (LAN)enp4s0 (LAN)

Some diagnostics output below:

root@qotom01:~# lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 39 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 4 On-line CPU(s) list: 0-3 Vendor ID: GenuineIntel BIOS Vendor ID: Intel(R) Corporation Model name: Intel(R) Atom(TM) CPU C3558R @ 2.40GHz BIOS Model name: Intel(R) Atom(TM) CPU C3558R @ 2.40GHz CPU @ 2.4GHz BIOS CPU family: 178 CPU family: 6 Model: 95 Thread(s) per core: 1 Core(s) per socket: 4 Socket(s): 1 Stepping: 1 CPU(s) scaling MHz: 52% CPU max MHz: 2400.0000 CPU min MHz: 800.0000 BogoMIPS: 4800.00 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology no nstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg cx16 x tpr pdcm sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave rdrand lahf_lm 3dnowprefetch cpuid_ fault epb cat_l2 ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust smep erms mpx rdt_a rdseed smap clflushopt intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves dtherm arat pln pts m d_clear arch_capabilities Virtualization features: Virtualization: VT-x Caches (sum of all): L1d: 96 KiB (4 instances) L1i: 128 KiB (4 instances) L2: 8 MiB (4 instances) NUMA: NUMA node(s): 1 NUMA node0 CPU(s): 0-3 Vulnerabilities: Gather data sampling: Not affected Itlb multihit: Not affected L1tf: Not affected Mds: Not affected Meltdown: Not affected Mmio stale data: Not affected Retbleed: Not affected Spec rstack overflow: Not affected Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Spectre v2: Mitigation; Retpolines, IBPB conditional, IBRS_FW, STIBP disabled, RSB filling, PBRSB-eIBRS Not affected Srbds: Not affected Tsx async abort: Not affected root@qotom01:~# lsusb Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub Bus 001 Device 003: ID 05e3:0608 Genesys Logic, Inc. Hub Bus 001 Device 002: ID 046d:c31c Logitech, Inc. Keyboard K120 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub root@qotom01:~# lspci 00:00.0 Host bridge: Intel Corporation Atom Processor C3000 Series System Agent (rev 11) 00:04.0 Host bridge: Intel Corporation Atom Processor C3000 Series Error Registers (rev 11) 00:05.0 Generic system peripheral [0807]: Intel Corporation Atom Processor C3000 Series Root Complex Event Collector (rev 11) 00:06.0 PCI bridge: Intel Corporation Atom Processor C3000 Series Integrated QAT Root Port (rev 11) 00:09.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #0 (rev 11) 00:0a.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #1 (rev 11) 00:0b.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #2 (rev 11) 00:0c.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #3 (rev 11) 00:0e.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #4 (rev 11) 00:0f.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #5 (rev 11) 00:10.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #6 (rev 11) 00:11.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #7 (rev 11) 00:12.0 System peripheral: Intel Corporation Atom Processor C3000 Series SMBus Contoller - Host (rev 11) 00:13.0 SATA controller: Intel Corporation Atom Processor C3000 Series SATA Controller 0 (rev 11) 00:14.0 SATA controller: Intel Corporation Atom Processor C3000 Series SATA Controller 1 (rev 11) 00:15.0 USB controller: Intel Corporation Atom Processor C3000 Series USB 3.0 xHCI Controller (rev 11) 00:16.0 PCI bridge: Intel Corporation Atom Processor C3000 Series Integrated LAN Root Port #0 (rev 11) 00:17.0 PCI bridge: Intel Corporation Atom Processor C3000 Series Integrated LAN Root Port #1 (rev 11) 00:18.0 Communication controller: Intel Corporation Atom Processor C3000 Series ME HECI 1 (rev 11) 00:1a.0 Serial controller: Intel Corporation Atom Processor C3000 Series HSUART Controller (rev 11) 00:1f.0 ISA bridge: Intel Corporation Atom Processor C3000 Series LPC or eSPI (rev 11) 00:1f.2 Memory controller: Intel Corporation Atom Processor C3000 Series Power Management Controller (rev 11) 00:1f.4 SMBus: Intel Corporation Atom Processor C3000 Series SMBus controller (rev 11) 00:1f.5 Serial bus controller: Intel Corporation Atom Processor C3000 Series SPI Controller (rev 11) 01:00.0 Co-processor: Intel Corporation Atom Processor C3000 Series QuickAssist Technology (rev 11) 02:00.0 Non-Volatile memory controller: Phison Electronics Corporation PS5013 E13 NVMe Controller (rev 01) 04:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 05:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 06:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 07:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 08:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 09:00.0 PCI bridge: ASPEED Technology, Inc. AST1150 PCI-to-PCI Bridge (rev 03) 0a:00.0 VGA compatible controller: ASPEED Technology, Inc. ASPEED Graphics Family (rev 30) 0b:00.0 Ethernet controller: Intel Corporation Ethernet Connection X553 10 GbE SFP+ (rev 11) 0b:00.1 Ethernet controller: Intel Corporation Ethernet Connection X553 10 GbE SFP+ (rev 11) 0c:00.0 Ethernet controller: Intel Corporation Ethernet Connection X553 Backplane (rev 11) 0c:00.1 Ethernet controller: Intel Corporation Ethernet Connection X553 Backplane (rev 11) root@qotom01:~# lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 39 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 4 On-line CPU(s) list: 0-3 Vendor ID: GenuineIntel BIOS Vendor ID: Intel(R) Corporation Model name: Intel(R) Atom(TM) CPU C3558R @ 2.40GHz BIOS Model name: Intel(R) Atom(TM) CPU C3558R @ 2.40GHz CPU @ 2.4GHz BIOS CPU family: 178 CPU family: 6 Model: 95 Thread(s) per core: 1 Core(s) per socket: 4 Socket(s): 1 Stepping: 1 CPU(s) scaling MHz: 52% CPU max MHz: 2400.0000 CPU min MHz: 800.0000 BogoMIPS: 4800.00 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology no nstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg cx16 x tpr pdcm sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave rdrand lahf_lm 3dnowprefetch cpuid_ fault epb cat_l2 ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust smep erms mpx rdt_a rdseed smap clflushopt intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves dtherm arat pln pts m d_clear arch_capabilities Virtualization features: Virtualization: VT-x Caches (sum of all): L1d: 96 KiB (4 instances) L1i: 128 KiB (4 instances) L2: 8 MiB (4 instances) NUMA: NUMA node(s): 1 NUMA node0 CPU(s): 0-3 Vulnerabilities: Gather data sampling: Not affected Itlb multihit: Not affected L1tf: Not affected Mds: Not affected Meltdown: Not affected Mmio stale data: Not affected Retbleed: Not affected Spec rstack overflow: Not affected Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Spectre v2: Mitigation; Retpolines, IBPB conditional, IBRS_FW, STIBP disabled, RSB filling, PBRSB-eIBRS Not affected Srbds: Not affected Tsx async abort: Not affected root@qotom01:~# lsusb Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub Bus 001 Device 003: ID 05e3:0608 Genesys Logic, Inc. Hub Bus 001 Device 002: ID 046d:c31c Logitech, Inc. Keyboard K120 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub root@qotom01:~# lspci 00:00.0 Host bridge: Intel Corporation Atom Processor C3000 Series System Agent (rev 11) 00:04.0 Host bridge: Intel Corporation Atom Processor C3000 Series Error Registers (rev 11) 00:05.0 Generic system peripheral [0807]: Intel Corporation Atom Processor C3000 Series Root Complex Event Collector (rev 11) 00:06.0 PCI bridge: Intel Corporation Atom Processor C3000 Series Integrated QAT Root Port (rev 11) 00:09.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #0 (rev 11) 00:0a.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #1 (rev 11) 00:0b.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #2 (rev 11) 00:0c.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #3 (rev 11) 00:0e.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #4 (rev 11) 00:0f.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #5 (rev 11) 00:10.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #6 (rev 11) 00:11.0 PCI bridge: Intel Corporation Atom Processor C3000 Series PCI Express Root Port #7 (rev 11) 00:12.0 System peripheral: Intel Corporation Atom Processor C3000 Series SMBus Contoller - Host (rev 11) 00:13.0 SATA controller: Intel Corporation Atom Processor C3000 Series SATA Controller 0 (rev 11) 00:14.0 SATA controller: Intel Corporation Atom Processor C3000 Series SATA Controller 1 (rev 11) 00:15.0 USB controller: Intel Corporation Atom Processor C3000 Series USB 3.0 xHCI Controller (rev 11) 00:16.0 PCI bridge: Intel Corporation Atom Processor C3000 Series Integrated LAN Root Port #0 (rev 11) 00:17.0 PCI bridge: Intel Corporation Atom Processor C3000 Series Integrated LAN Root Port #1 (rev 11) 00:18.0 Communication controller: Intel Corporation Atom Processor C3000 Series ME HECI 1 (rev 11) 00:1a.0 Serial controller: Intel Corporation Atom Processor C3000 Series HSUART Controller (rev 11) 00:1f.0 ISA bridge: Intel Corporation Atom Processor C3000 Series LPC or eSPI (rev 11) 00:1f.2 Memory controller: Intel Corporation Atom Processor C3000 Series Power Management Controller (rev 11) 00:1f.4 SMBus: Intel Corporation Atom Processor C3000 Series SMBus controller (rev 11) 00:1f.5 Serial bus controller: Intel Corporation Atom Processor C3000 Series SPI Controller (rev 11) 01:00.0 Co-processor: Intel Corporation Atom Processor C3000 Series QuickAssist Technology (rev 11) 02:00.0 Non-Volatile memory controller: Phison Electronics Corporation PS5013 E13 NVMe Controller (rev 01) 04:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 05:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 06:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 07:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 08:00.0 Ethernet controller: Intel Corporation Ethernet Controller I225-V (rev 03) 09:00.0 PCI bridge: ASPEED Technology, Inc. AST1150 PCI-to-PCI Bridge (rev 03) 0a:00.0 VGA compatible controller: ASPEED Technology, Inc. ASPEED Graphics Family (rev 30) 0b:00.0 Ethernet controller: Intel Corporation Ethernet Connection X553 10 GbE SFP+ (rev 11) 0b:00.1 Ethernet controller: Intel Corporation Ethernet Connection X553 10 GbE SFP+ (rev 11) 0c:00.0 Ethernet controller: Intel Corporation Ethernet Connection X553 Backplane (rev 11) 0c:00.1 Ethernet controller: Intel Corporation Ethernet Connection X553 Backplane (rev 11) root@qotom01:~# ip link 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: enp4s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:76 brd ff:ff:ff:ff:ff:ff 3: enp5s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:77 brd ff:ff:ff:ff:ff:ff 4: enp6s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:78 brd ff:ff:ff:ff:ff:ff 5: enp7s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:79 brd ff:ff:ff:ff:ff:ff 6: enp8s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:7a brd ff:ff:ff:ff:ff:ff 7: eno1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:7b brd ff:ff:ff:ff:ff:ff altname enp11s0f0 8: eno2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:7c brd ff:ff:ff:ff:ff:ff altname enp11s0f1 9: eno3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:7d brd ff:ff:ff:ff:ff:ff altname enp12s0f0 10: eno4: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode DEFAULT group default qlen 1000 link/ether 20:7c:14:f2:9c:7e brd ff:ff:ff:ff:ff:ff altname enp12s0f1

Zooming in on remote education and WebRTC

bloggeek - Mon, 11/06/2023 - 12:30

An overview of remote education and WebRTC. The market niches, challenges and solutions.

Whenever a video meetings company starts looking at verticals for the purpose of targeted marketing, one of the verticals that is always there is education. We’ve seen this during the pandemic – as the world went into quarantine mode, schools started figuring out how to teach kids remotely.

The remote education market is not just schools doing remote video calls. It is a lot more varied. I’d like to explore that market in this article.

Table of contents

How big can remote education really get?
Me? Remote education?
The role of WebRTC in remote education
Top down decisions; sometimes
Live, online and in-person
Hybrid
Moderation
Assessment
Collaboration and whiteboarding
The challenge of engagement
The challenge of engagement (part 2)
Asymmetry in remote education
Training the educators
A matter of costs

How big can remote education really get?

There are around 2 billion children in the world. Over 80% of them attend schools.

Some 235 million higher education students are out there as well around the globe.

During the pandemic, a lot of them were online, taking classes remotely. For multiple hours each day.

The slide above is from Kranky Geek 2020. In this session, Google talked about their work on WebRTC in Chrome.

Here they shared the increase in video minutes during the initial quarantines. The huge spike there starts at around the August/September timeframe, when schools start.

Remote education is here to stay. Not with its increased usage of 10-100x, but definitely bigger than in the past. There are many places where remote education can fit – and not only for emergencies such as the pandemic.

Me? Remote education?

Like everyone else, my kids went through the process of remote education during the pandemic. Here, the Ministry of Education went all-in with Zoom for schools (along with Google Classroom and Microsoft Office – go figure). Since then, our kids have on and off private tutors doing classes remotely sometimes. And now, when we have a war raging between Gaza and Israel, depending on where you live, you might be studying from home or physically in school.

I had my share of consulting with education organizations across the globe. Some focusing on schools, others with universities and some with private tutoring. It was always fascinating to see how such markets are distinctly different from each other, and how remote education also takes different shapes and sizes based on the country.

And then there are my own online courses, with their associated office hours and AMAs.

The role of WebRTC in remote education

WebRTC plays an important role in the education market. Besides offering video communications, it also enables the ability to mesh the communication experience directly into the LMS (Learning Management System) or the SIS (School Information System), offering a seamless and tailored experience for both the teacher and the learners – one that enables the educators to implement various pedagogies.

Remember here that WebRTC is a synchronous technology – live, real-time voice and video communications. A large chunk of the education market is leaning heavily on asynchronous learning (recorded videos, texts to read, etc). These are not covered in this article.

Here are some market niches and use cases where you will find WebRTC in remote education.

Group lessons

The simplest one to explain is probably group lessons. The classic one would be the pandemic use case, where during quarantine, schools went all virtual – classes were conducted online.

Remote group lessons aren’t limited to schools either – they are done in universities, private group tutoring, etc.

Main challenges here include:

Moderation tools for the teachers. Ones that are simple to use while conducting the lesson itself

Collaboration tools to make the lessons more engaging. Maintaining engagement in online group lessons is the biggest challenge at the moment, especially for younger learners

Authentication and authorization of users. Lots of anecdotal stories around this one throughout the pandemic

One thing that is raised time and again with group lessons, especially in schools, is the need (and inability) to get the students to keep their cameras on. This is a huge obstacle to effective learning, and something that needs to be taken into account.

Another important thing that needs to be fleshed out early on here, is who is the client – is it the teacher or the students. Whoever the system is geared towards will set the tone to how the solution gets designed and implemented.

One-to-one tutoring

These are mainly one on one lessons conducted remotely.

Outside of the domain of classic education, a lot of classes are actually conducted in such a way. Here are a few anecdotal stories from recent years that I’ve learned about:

A dear friend who is learning to play the piano. Remotely. She travels a lot between the US and Israel, and takes her lessons from everywhere through her iPad

Another friend, taking 1:1 drawing lessons

Online chess lessons for kids in our community

My son’s friend, learning C++ on Unreal engine, taking 1:1 lessons

My son, a few years ago, when he was 10 or so, learning to build online games using nocode game engines from an 18 year-old who lived two cities away

My wife took online dance lessons to specialize in Salsa from a renowned instructor abroad

Besides the collaborative, engagement level and nature of such lessons, it is important to note that they aren’t suitable for everyone. Some teachers are more natural in these, and some students can learn effectively in such a manner while others struggle (I have both examples at home).

An interesting use case here that I’ve seen is math and English (!) tutors from India and China teaching remote kids in the UK and the US. Why? Simply because they are cheaper than using local teachers. Then there was the opposite – rich Chinese families getting one-to-one English tutoring for their kids from US teachers. Go figure.

One-to-one tutoring comes in a lot of different shapes and sizes.

MOOCs (Massive Open Online Courses)

MOOCs were all the rage 10 years ago. Their market is still consistently growing.

MOOCs are simply large online courses that are open for people around the globe. Some of them are collaborative, while others are mainly lecturer driven. Some allow for asynchronous learning while others are more synchronous in their nature. Both the asynchronous and synchronous learning modes in MOOCs offer self-paced learning (at least to some degree).

WebRTC finds its way into MOOCs for their synchronous part, when that requires live video sessions – either between lecturers and students or between student groups in the more collaborative courses.

Proctoring

Proctoring isn’t about learning, but about taking exams. Remote proctoring enables taking exams at the comfort of one’s home or office without going to the classroom.

With proctoring, the user is required to open up his camera and microphone as well as share his screen while taking the exam. The proctoring application takes care of checking that other tabs aren’t being opened and that nothing fishy is taking place (as much as possible). WebRTC is used to gather all that realtime audio and video data and record it. If needed, these recordings can be accessed by human proctors later on.

It should be noted that for proctoring, there are a lot of requirements around circumventing the ability to cheat on the exam. This includes things like monitoring applications used during the exam, maintaining focus on the exam page, etc. To achieve this, most proctoring solutions end up as PC applications (usually using Electron) which the student needs to install on his machine in order to take the exam. The innards of the proctoring application will end up using WebRTC in a web application – simply for its speed of development and the use of the WebRTC ecosystem.

Coaching

While similar to classic education, coaching is slightly different. In its essence, these can be 1:1 sessions or small group sessions where issues and challenges in certain areas get fleshed out. In group lessons and 1:1 tutoring, a lot of the focus is on collaboration features. Here, in many cases, it will be more on the video of the participants and the need to bring them together.

Another interesting aspect of coaching is the platform it gets attached to – either directly or indirectly. Coaching often comes bundled as a larger course/training offering, mixed with in-person meetings, reading/presented materials and the coaching sessions themselves.

The LMS and SIS systems are usually also lacking in the coaching platforms. Usually, these will be geared towards flexible use and at times an integrated payment system.

Webinars

Webinars are a form of lessons that is conducted over the internet, mostly for businesses to assist in marketing and sales efforts. Depending on the level of interactiveness of the webinar, the need and use of WebRTC will be needed.

In the past, webinars were usually conducted via specialized downloadable applications, where the content was mostly slide decks and the voice of the speakers. The interaction with the audience was done via text messages and organized Q&A. Over time, these solutions became richer and more sophisticated, adding video communications as well as the ability of the audience to “join the podium” if and when needed.

Using WebRTC here enabled getting rid of the application download requirement and increased the level of interactivity quite considerably.

The intersection of education and healthcare

Education and healthcare are bound together. I’ve shown that a bit in my WebRTC in telehealth article, looking at it from the remote training of healthcare topics perspectives. I want to take a different angle on the same topic here. I’ll do that by showcasing two interesting use cases I’ve been privy to a few years back.

#1 – Dance lessons in cancer

I heard this one from a dancer who had cancer and healed. Women with cancer have it hard. Chemo is brutal – it seeps out the energy and causes hair loss. This means women don’t want to go outside that much. Here, being able to bring them remotely to a dance lesson can be a real benefit to them, especially if they love(d) dancing. They won’t go physically – not wanting to meet people outside and the stairs that come with it – along with the energy it takes. But they will be willing to dance – maybe.

Remote dance lessons for this niche is beneficial. Not from an educational standpoint but more from a mental health one.

#2 – Video in class for students in hospitals

Another vendor I worked with briefly was assisting school kids who had to be treated in hospital or just stay home for prolonged periods of time (think weeks or months at a time). Their solution was to bring a video conferencing system and rig it in the physical classroom of the kid as well as where he is located, be it home or a hospital bed.

This way, the kid could join the classes as well as stay connected to other classmates during recesses. The main purpose here isn’t really the teaching part, but rather to make sure the student stays in contact with peers in his age group and not be secluded during that period of time.

Is this a use case in education? In healthcare? I can’t really say…

ERT (Emergency Remote Teaching)

The pandemic showed us that remote education is challenging but might be necessary. We were all quarantined for long periods of time, with school across the globe going remote.

Here in Israel, when clashes with Gaza or Hezbollah in Lebanon flare, schools shift to remote learning. It isn’t frictionless or smooth, but it is the solution we have to try and continue educating kids here.

The most crucial aspect of ERT is that teachers are forced to change their teaching setting with no preparation. In Israel, at least, the pandemic didn’t prepare teachers for the current war – it feels like the education system in Israel learned nothing from the pandemic wrt to remote teaching

Top down decisions; sometimes

Education is interesting. Especially the institutional ones of schools.

In some countries, decisions are made top down while in others, there’s more autonomy kept at the school level or the district level.

Here are a few things I learned asking the question on LinkedIn, about what tool was used during the pandemic for virtual classes across the globe:

Israel. Where I live. Was mostly Zoom during the pandemic
- There was also a bit of Google Meet and some BigBlueButton, due to its integration with MASHOV (an SIS in Israel)
- The government struck a deal in education for Google Classroom country-wide
- There’s also Office available for free for all students
- And Zoom was the decision for virtual classes
- This year, it all changed to Google Meet, presumably due to security concerns, but more likely this was due to pricing (Zoom renewal cost money while you get Google Meet and Microsoft Teams for free with Google Classroom and Office respectively)
- Zoom hurried up with a statement that it is secure and now available for free for the education system in Israel
- As the saying goes – it’s all about the money
Bulgaria used Jitsi Meet (through the Shkolo platform); later replaced by Microsoft Teams. Both with government provided accounts
Colombia. Most public schools and the university system relied heavily on Microsoft Teams. Private schools and universities were about an even split between Zoom and Microsoft Teams
Austria was mainly Microsoft Teams
Russia – Zoom
The United States was mostly Zoom. It wasn’t mandated, but just how things ended up in most places
UK. A private school in London opted for Microsoft Teams. Public schools were left to figure out their own solution
Argentina. Zoom, though I am not sure if everywhere and if the decision was top down or bottom up
India. Primarily Microsoft Teams and occasionally Zoom. Mainly because Microsoft Teams had better and stronger channel partners in India, being able to offer better deals
France. Started with Zoom and Jitsi Meet in schools. Now, the government has built a large scale BigBlueButton infrastructure for virtual classrooms

This is by no means complete or accurate, but it shows a few important aspects of education:

In some countries, decisions on the tools to use is taken top down, while in others, each district or school is left to autonomously make a decision

Like in many industries, but probably more so, appearances matter. Losing Israel for Zoom was bad publicity. They had to fix that quickly by renewing the service for free. BTW – the damage is already done, my kids are now using Google Meet at school and there likely isn’t a way back

Live, online and in-person

Education is mixed. It isn’t all virtual and isn’t all in person.

My own WebRTC Courses are online, but not live. The lessons are pre-recorded. I offer monthly AMA meetings as part of them which are online and live.

I took a CPO course last year. It included in person meetings (3 full days), weekly live sessions as well as pre-recorded information.

My kids are now learning some days remote and some days in-person in the school.

Some countries had recorded/broadcasted lessons alongside virtual live classes during the pandemic, creating from them a full set of learning materials that students can use moving forward.

The LMS (Learning Management System) used needs to take all these into account, enabling different learning strategies and different content types. Your own service needs to be able to figure out what works best.

Hybrid

The term Hybrid Learning refers to any form that incorporates online and offline learning. This is slightly different from how we define hybrid meetings.

As an example in Israel at the moment, in the current “war setup”, students go physically to school a few days a week and the rest they learn asynchronously or synchronously from home.
Another example of hybrid learning is when students work with laptops in the traditional classroom.

Allowing a student to join remotely to a class taking place in-person is a real challenge, but one that needs to be dealt with as well. This isn’t any different from hybrid meetings in enterprises in terms of the basic need. The difference is likely in size and complexity.

Most classes aren’t geared to this. From the placement of the cameras in the class, to the way the lessons are conducted and to the way teachers need to split their attention between in person to remote students.

In most places, going hybrid in education is an intentional decision that can be made only for select use cases and in a limited number and types of institutions.

Moderation

Who is allowed to join a virtual lesson? Should the teacher approve each student joining? How do you know who is online? Who is actively listening? Should anyone be automatically allowed to speak up? Share their screen? Is there a way to check if the student goes “off the reservation”, doing other things in other browser tabs or on his phone in parallel?

All these are hard questions with no good answers.

Moderation in education must take place – especially for group lessons. This has two purposes:

Maintain a semblance of order
Let the teacher focus on teaching

Oftentimes, moderation tools deal with a semblance of order but less with the focus of the teacher or teaching.

The decision in Israel for example to go for Google Meet makes total sense simply because authentication and identity is managed by Google Classroom already. Classroom is acting as the LMS as well, or at least the hub for students and teachers. Having a tighter integration means some of the moderation requirements can more easily be met.

It isn’t only about what can be moderated, but how and with what level of friction

Assessment

How are assessments taking place in online learning?

In the traditional classroom, teachers physically saw the students and could easily gauge their level of attentiveness. To that, home assignments and tests were added.

Once going online, technology can come to assist the teachers and students, adding a layer of information to the assessment process. Dashboards can be built to make this data accessible.

Where does WebRTC fit in here? The same way it does in online meetings, where we see today a growing focus on incorporating transcriptions, meeting summaries and action items automatically. Similar LLM/generative AI technologies can be used to glean insights out of online lessons.

In many ways, this isn’t done yet. Probably because we’re still struggling with engagement (see below).

Collaboration and whiteboarding

How is collaboration done in education? Do we need the classing blackboard/whiteboard for teaching? How does that get translated to the digital, remote scenario?

Are we looking here for something as powerful and flexible as a Miro board or something simpler and less feature rich?

Is teaching math or physics similar to teaching languages or literature when it comes to collaboration and whiteboard?

How about Kahoot or similar polling/quiz capabilities? Do we make them engaging or boring as hell?

A lot of thought and energy needs to be diverted towards these types of questions, in trying to figure out what works best to increase engagement and improve the learning experience (and by extension, the learning itself).

The challenge of engagement

How do you define engagement in online synchronous lessons?

Is students opening cameras considered engagement?

Maybe students be engaged with their cameras turned off

Getting students to open up their cameras, having them choose to do so and keep the cameras on is a big issue in schools and in higher education.

In my son’s school, they are now shifting towards enforcing students to open their cameras… but allowing them to point that camera at the ceiling

Once you have cameras on, how does a teacher gauge the level of engagement of a student? How does he spare the time looking at 20+ students (36 in Israel classes) to understand if they are engaged or not while trying to present his screen to teach something out of his slidedeck?

“Feeling the crowd” to understand if a topic needs further explanation or can the teacher move on to new topics is harder to achieve online than it is in person.

The challenge of engagement (part 2)

How do you get students engaged?

What type of collaboration solution do you need?

Which experiences should be baked into the solution?

My son decided to take up Russian. His friend speaks Russian with his parents, so he decided he wants to understand when they talk to each other (go figure). He decided independently to install Duolingo on his phone and has been taking their lessons for almost a year now

He can now read Russian and know quite a few words.

A good friend of mine is learning German using Duolingo. We did a roadtrip in the US in February. I had to hear him learn in our long hours on the road. It was an interesting experience to see it from the side, trying to figure out how this magic happens.

Engagement and “gamification” are a main part of how Duolingo works and how it gets students back into their app over and over again.

We haven’t quite cracked the formula of how to do this well in live virtual classes. There must be a way to get there, and when we find it, we will see great dividends from it.

Asymmetry in remote education

There are teachers and there are students. Who is the system designed to cater?

A simple question. Answering with “both” is likely going to be wrong most of the time.

I had a meeting at a large and prominent university in Europe a few years back. They wanted to build a video conferencing system for lectures. Have the professor in front of a large digital board showing tens of students joining remotely. Call it extremely expensive and unique. That was before the pandemic, so unrelated to it.

The question I had was who this system is for. Is it to sell students on a great remote experience or is it for the professor to feel important. I have my own answer here

You need to decide who the service you are developing is really there to cater – the teacher and his needs, assuming that students will simply join because they have little choice. Or the students, focusing on enticing them to join, collaborate and interact.

Doing both at the same time is a real challenge, and one that most vendors aren’t prepared to take yet.

Figure out who your main user is. The teacher or the students. Or maybe the parents?

Training the educators

Someone needs to teach the teachers how to use the service. This is a real problem, especially when going mainstream.

When the pandemic started and Zoom was selected here in Israel, a lot of videos surfaced explaining how to use Zoom in the context of teaching with it. Last month, when Google Meet was the official solution, you started seeing the same occur for Google Meet here in Israel.

The differences between these two services may seem minor, but they are big for teachers who aren’t technically savvy.

Some private tutors for example shy away from remote lessons. Their reason is the inability to focus on the student during the lesson. Increase that by 20-40 students in a single lesson, many of them acting like prisoners trying to break out and figuring out ways to game the system called a virtual lesson, and you get to the need for teachers who know their way using the service inside and out.

Onboarding and familiarizing teachers to the platform is just as important as the actual service, sometimes even more

A matter of costs

This one might just be an opinion of mine.

Remote education is a huge market. During the pandemic, it encompassed almost all the world’s students. And yet, the amount of money available to spend per minute is quite low.

In many cases, the deals are large (in front of a state or a country). Sometimes, they are smallish, in front of a single school. There’s money in these institutions, but in many cases, that money is spent elsewhere.

When going after the education market, it is vital to understand the buying habits and budget of the would-be purchaser beforehand.

Solutions in the education market need to be cost effective and efficient from a WebRTC infrastructure point of view

Where can I help, if at all?

Online WebRTC courses, to skill up engineers on this technology

Consulting, mostly around architecture decisions and technology stack selection

Testing and monitoring WebRTC systems, via my role as Senior Director at Cyara (and the co-founder of testRTC)

The post Zooming in on remote education and WebRTC appeared first on BlogGeek.me.

WebRTC in telehealth: More than just HIPAA compliance

bloggeek - Mon, 10/23/2023 - 13:00

When it comes to WebRTC in telehealth, there are quite a few use cases and a lot of things to consider besides HIPAA compliance.

A thing that comes up in each and every discussion related to telehealth & WebRTC is the value of the call in telehealth. We’ve seen video meetings and calls go down to zero in their cost/value for the user. Especially during the pandemic. So whenever we find a nice market where there is high value for a call, it is heartening. Healthcare is such a place where we can easily explain why calls are important.

But what exactly does WebRTC in telehealth mean? It isn’t just a patient calling a doctor. There is a lot more to it than that. Let’s dive in together to see what we can find.

Table of contents

My own experience with Telehealth
- As a user
- As a consultant and a product manager
Finding WebRTC in Telehealth
A game of numbers
WebRTC telehealth and HIPAA compliance
Network and firewall restrictions
Quality of media
Asymmetric nature of users and devices
Medical devices, sensors and telemetry
SaaS, CPaaS & open source: Build vs Buy

My own experience with Telehealth As a user Me and my son, waiting in a hospital while he had some blood samples taken during COVID

Like many others, my first real bump with telehealth took place during the COVID quarantines.

My son was sick with high fever for over a week, and the doctors didn’t help any.

My wife was worried, needing more comfort by knowing someone was looking at him. Really looking at him.

So we used a kind of a private service that a hospital near our vicinity was giving:

You subscribe and pay a hefty price
They send over a kit
You install an app and take measurements multiple times a day (useless ones, but stay with me)
They send over a radiologist to do an X-ray scan (need something to show they can)
Then you get to talk to a doctor once a day. Over a video call. From the same app

What can I say? It worked as advertised.

As a consultant and a product manager

We have quite a few healthcare clients using our various WebRTC services at testRTC.

Other than that:

Took part of an RFP of the ministry of health in Israel by assisting the vendor who approached me win the contract
I assisted vendors during the pandemic to troubleshoot their architecture and scale their service rapidly

That and just from conversations with vendors, along with a review of this article by a few who work on telehealth products and integrating their comments as well.

Does that make me an expert in telehealth? No.

But I can fill in the WebRTC angle of telehealth, which is a rather big one.

Finding WebRTC in Telehealth

Telehealth for me is about the digital transformation of healthcare services.

It can start small, with things such as scheduling and viewing lab test results. And then it can grow towards virtualizing the actual patient-doctor interaction. Or any other interaction within the healthcare space between one or more people (emphasis on one here – not two).

I’ve listed here the main use cases that came to mind thinking of it in recent days.

Patients and doctors

The most obvious use case is the patient and doctor scenario.

In this, the doctor visitation itself is remote and virtual.

This can be useful in many situations:

When the patient can’t get to the doctor’s office
During the pandemic:
- When healthcare providers didn’t want patients physically in the office
- If doctors are sick, but their numbers are dwindling due to them being quarantined, while they can still be useful as doctors remotely
If you don’t want to waste a patient’s time in coming over and waiting
When it is truly urgent (an emergency)

For many of these situations, this is the setup that takes place:

Doctor – sitting in front of a PC or laptop. In a designated office or hospital (=managed network), or at home (=unmanaged network)
Patient – connecting from a smartphone or tablet, via a direct link or an installed application

No. I am not ok

bloggeek - Mon, 10/09/2023 - 07:15

I’ve been meaning to write about a different topic about WebRTC, but somehow, this was more important.

There’s a war going on here where I live between Israel and Hamas. Or Israel and Gaza. Or Israel and the Palestiniens. Or Israel and Iran’s proxies. Or Israel and muslim extremists.

Or all of the above if we’re frank with ourselves.

We haven’t invited this war or wanted it, but it is what we need to face and deal with.

Others are explaining the situation better than I can on social media sites and in english. Here is one such example:

To those of you who reached out to me asking if I am ok, if me and my family are safe, I answered that we’re ok’ish mostly.

Well… I am not ok.

At least 700 were brutally murdered
- Many of them civilians
- Many of them babies, children, women and the elderly
- Some of them are muslims (usually through rockets)
- Some of them foreigners here in Israel – working, living or just visiting
Over 260 were butchered in an outdoor party. Many of them teenagers and young adults
The number of murdered is likely to raise above 1,000
- At the size of Israel, this is bigger than 9/11 or pearl harbor event
- It is a huge milestone and likely a turning point
Over 5,000 rockets have been fired on Israeli cities (might be more – might be less – who’s counting anymore?)
There are more than 100 kidnapped Israelis in Gaza now. Taken from their homes in Israel. Again – babies, children, women and elderly among them
Israeli parents and families still don’t know where their loved one are
- Are they wounded somewhere?
- Are they dead?
- Were they kidnapped and taken into Gaza?
- Are they being abused? Raped? Decapitated?
- Some find out from social media
  - A story about people seeing their family members on live videos
  - An elderly woman whose family found out she was murdered because the murderer decided to take a photo of her and publish it on her Facebook account
- My Facebook is filled with photos of missing family members. Mostly kids and young adults
This is all for the world to see right there on social media if one cares to look at war crimes and atrocities committed by Hamas while the Gazans, Palestinitens and other extremist muslims across the globe cheer and gloat (again – directly on social media – just go and watch)
These aren’t human beings. These are monsters

I. Am. Not. Ok.

Yes. Physically, I am fine. We live at the center of Israel in relative safety at the moment
Everything is relative in life
We came back from a two week vacation in the US a day before the war started
Yesterday, I went to the supermarket to buy supplies – we’re short on everything
In the elevator I met a neighbor coming out. We greeted each other with “hi”. He noted that we don’t say “good morning” anymore. I agreed. We left without the so common “have a great day” greeting
The supermarket was big and full of people for a Sunday morning
It was also totally quiet. If you know Israelis, you know we’re a loud bunch. None of it took place there
Everyone looked shell shocked and subdued on the outside. Looking more closely, you could see purpose. A parent telling his 20-year old child he wants to be called to the war – saying that while he is old, he wants to participate and help in any way he can
A person near the cash registers, asking people to donate food and stuff to take to the soldiers
And me? I consider myself sharp minded and grounded. I couldn’t find my shopping cart each time I went hunting for things to buy. Over and over again. I even came back and almost took a different cart to the astonishment of the pregnant lady and her husband standing next to it. Where was my mind wondering? Each and every time

No. I am not ok.

We have two kids. Teenagers
My son was on overdrive on the first day of the war. Hyperactive
- Probably in an attempt to process things
- That curbed down by the end of the day, and now he is silent and subdued
- Buries himself in his video games and his drawings
My daughter, ever the silent type, stayed silent
- She went to sleep on that first day, telling me that one of her best friend’s brothers was likely injured and his parents are rushing to the hospital
- She woke up the next morning reading his name on a website as one of the first people announced dead. Murdered. Only 19 years old
- Before we could tell her the news as we heard it through the parents
- She spent the rest of yesterday going back and forth with the rest of her friends to that friend’s home. She will likely do that the rest of the week
- At the age of 16, she is now experiencing grief. Seeing it in the face. Seeing parents bury their murdered child
- What can I do with such reality?
And me and my wife? We trudge along, each with his own way of dealing with it
Thinking if and when to do what
- Is it the right time to shower or should we wait? Sirens and all
- Should we take our kids to this activity or that, or just cancel it for now
- And if our kids need to go somewhere, should we go along with them, for the good that will do, or not
- Is it enough to just close the door to the Mamad, or do we need to add an element that won’t let murdering palestinians open it from the outside while we’re inside?
- Mundane daily thoughts and decisions we need to make here
It is hard to sleep at night
- Not sure it has anything to do with a jet lag coming back from the US and the 10 hour difference
- Or is it just the weird situation we’re in
- Probably that second option

I am not ok.

We had sirens here. 5 of them so far I think. Not really counting
Each time, this means running to our Mamad. Every house and apartment in Israel has such a thing if it were built in the last 20+ years
- This is a room that is built differently than the rest of the house
- It has concrete walls and ceiling
- A bomb shelter door and window made of heavy iron
- Complete with the ability to ceil it up for chemical weapons if needed
This room is also my home office. If you’ve seen any of my videos or met me virtually, then you’ve seen this room
The window there is now closed. There’s no point in opening it up until this is all over
Once, I had to run in from a neighbor’s apartment, where we discussed matters related to the building. A decision I had to make – should I go stay in their Mamad or run home to be with my family so they worry a wee bit less
We had a rocket fall a few 100’s of meters from our place. On the road. No one was wounded. We heard it really well

I am not ok.

There’s an iron dome battery somewhere close. A few kilometers away I assume
When it fires rockets we feel it and then we hear it
It might be followed with a siren or not, depending on where the likely missiles are about to hit
Then you hear the intercepts or the falling missiles. They sound different

I am not ok.

We live next to a hospital. It is located some 2 kilometers from our place
In the last two days, I’ve seen my share of military helicopters coming in and out, moving severely wounded people around as they spread them across hospitals in Israel

I am not ok.

Hamas and the Palestinians are busy killing as many jews as they can indiscriminately
Our government and legal system are bickering over the legality of stopping supplying electricity to Gaza. We give them life while they give us death
What stupid world are we living in?

Physically? I am fine.

The rest? Not so much

–

If you know me or have been to this site before, then you know a bit about Israelis already.

We are here to create and innovate. To bring good to the world and to improve things.

In the 10+ years I’ve been running this blog, I shared my thoughts and helped my industry as much as I could. Many times, not asking for anything in return. It is what I do.

Two years ago, me and my other Israeli co-founders sold testRTC. Ever since I’ve been asking myself what I should do next.

One of my dreams recently has been to start teaching. Kids. Older ones. Show them the world of technology and entrepreneurship and what is possible. Be a mentor. Raise the next generation of creativity and innovation of Israelis.

I believe Israelis are a net positive to the world.

I act like this every day. I teach my kids in that way. I see that the floundering and ill equipped education system we have here in Israel does the same. There is no hatred in our teachings or in the way we raise our kids.

–

Palestiniens. Hamas. Extremist muslims.

How can they slaughter kids in cold blood? Murder whole families? Kill without discrimination whole communities? Then go and show it to the world on social media. And then praise it and celebrate on the streets.

This is inhumane.

In many ways, I see them as a net negative to the world.

I just can’t see it otherwise at the moment.

–

People who ask me what they can do to help – nothing. And everything.

Our dysfunctional government will find a way to help, and until then, the civilians here and the soldiers will figure it out. We always do. We don’t have a choice.

I don’t really need anything from you. We’re Israelis. We’ll survive. We have done so ever since the holocaust and we know we can only depend on ourselves. So thanks for asking, but I don’t need a thing at the moment.

The solidarity flags and colors lighting places across the globe? That’s useless. Sorry
You’ll switch gears over there saying we shouldn’t kill Palestiniens soon enough
All the while having your governments (at least some of them) continue to fund the Palestiniens in one way or another, just ending up fueling their war against us

Here’s a few picks from the news:

UK: Nir Bitton hits out at ‘brainwashed’ Celtic fans’ Palestine banners
Germany: German citizens are celebrating the butchering and kidnapping of Jewish people in Israel
US: Yosef Haddad, on The New York Times headlines (He is an Israeli Arab Christian. Follow him if you are interested in the Israeli-Palestinian conflict)

What can you do?

Understand that there aren’t really two sides to this story.

This conflict isn’t symmetrical in any way. It is between people who want to live and people who want to kill and ruin.

If you don’t believe me, then just go on social media and see what the Palestinians are doing. How they parade dead Israeli soldiers, small kids and elderly on the streets of Gaza for all their people to see and enjoy. This is the 21st century.

So no. I am not ok.

We will prevail. And in the meantime, I will be working. Different than usual, but still working. Still making my small and modest contribution to the world. Trying to touch and better those I interact with.

The post No. I am not ok appeared first on BlogGeek.me.

Fitting WebRTC in the brave new world of webcams, security, surveillance and visual intelligence

bloggeek - Tue, 09/26/2023 - 12:30

WebRTC has its place in surveillance and security applications. It isn’t core to these industries, but it is critical in many deployments.

Surveillance has become near and dear to my heart. I had a few vendors consult with me in the past. There are a few using testRTC. And then there’s the personal level. The system we have in our apartment building.

This got me to think quite a lot about WebRTC in surveillance tech lately.

Table of contents

Why my interest in surveillance cameras (and WebRTC)?
Security and surveillance use cases in WebRTC
Unique (?) challenges for WebRTC with camera hardware
Ingress, egress and the concept of real time
Mobile or desktop?
The age of Artificial Intelligence in surveillance tech
WebRTC isn’t core for surveillance but it is critical

Why my interest in surveillance cameras (and WebRTC)?

I live in an apartment building here in Israel:

23 floors

91 apartments

2 main entrances (and another side one)

3 elevators

3 levels of underground parking

…

And yes. We have a surveillance camera system. Like all of the other apartment buildings in my neighborhood:

The view from my apartment on a nice day

A year ago, I was in charge of the vendor selection and upgrade process of our cameras. We switched from an analog system into a hybrid analog/IP one.

This month, we’re looking into upgrading an elevator camera to an IP one, as well as adding WiFi to our underground parking. Having a chat with one of the vendors we’re reaching out to, he was fascinated with my work on WebRTC and the potential of using it for application-less viewing of cameras.

I’ve had my share of meetings and dealings with vendors building different types of surveillance and security solutions. From private security solutions to large scale, enterprise visual intelligence ones. Obviously, the matter of these interactions were around WebRTC.

I am not an expert in surveillance, so take the market overview with a grain of salt

That said, I do know my way with WebRTC and where it fits nicely

Here are some of the things I learned over the years

Security and surveillance use cases in WebRTC

I’ll start with the obvious – cameras, security and surveillance have multiple use cases. Some of them can be seen as classic to this domain while others slightly newer or a specialized niche. Each of these use cases is a world onto its own with its requirements from WebRTC and the types of solutions emerging in it.

Small scale / cheap multiple surveillance cameras

This is where I’d frame my own experience of our apartment building. A system that requires 32 or less video cameras, spread across the location, connected to a DVR (Digital Video Recorder) or an NVR (Network Video Recorder).

In essence, you go install the cameras in sensitive locations, wire them up (with an analog cable, IP or even wireless) to the media server that is located onsite as well. That media server is a DVR if it is a closed loop system or an NVR if you’re living in modern times. I’ll just refer to these two as xVR from here on.

Once there, you hook’em up to a local monitor that nobody goes and look at, as well as let the owner connect remotely from his PC or mobile phone.

Is WebRTC needed here? Not really.

Surveillance cameras today use RTP (and sometimes also RTSP). These are the new ones. Old ones are pure analog. They connect to that xVR media server, which handles them quite well today. It did so also before WebRTC came to our lives. The user then accesses the system to play the videos remotely using a dedicated application, which again, existed before WebRTC.

Since there’s no specific requirement to access this through a web browser, the use of WebRTC here is questionable.

You might say WebRTC would make things easier, but hey – if it ain’t broken, don’t fix it

These solutions are purchased from local vendors that install such systems. The buyer will usually reach out to an installer that will pick and choose the cameras and the surveillance system for the buyer. The buyer cares less about the technology and more about the local vendor’s ability to install and maintain the system when needed.

Enterprise / large scale surveillance

Large scale surveillance systems for enterprises is more of the same as the small scale ones, but with a few main differences:

There are more cameras
There are also more sensors which we want to control and manage, likely using the same system. Think doors and managing employee entrance using keycards for example. While this is about surveillance and security, it is also about building automation
This can go from a small scale building to as large as smart cities with lots of cameras – anywhere in-between that I bunch here are most likely multiple different markets with slightly different requirements
We are likely to have a NOC, where security guards look at screens. Just like in the movies…

The two things that are making headways in this industry?

Using AI to reduce the amount of people needed to look at surveillance monitors. This is done by adding vision smarts into cameras and the media servers (local or in the cloud), so that events and alerts can be filtered better
To some extent, there’s also a requirement to use WebRTC in the NOC to be able to view in real time camera feeds without installing anything

Like the small scale solutions, here too the buyer will look for local installers. These will be the local integrators who bring the systems and install them. At times, the decision of brand will come from the buyer, though this is less likely. It is important to remember that a considerable part of the cost goes towards the setup and installation and not necessarily to the cost of the equipment itself.

Personal/home surveillance

This one is the residential one. It is a B2C space where the buyer is a person buying a camera for his own home security. The decision is made on price or brand mostly.

Here you’ll find also solutions that make use of old smartphones and tablets as cameras, or something like the one we purchased a few years back when our kids were younger:

A digital peephole camera

Having the ability for them to see who is outside our door when they were shorter.

Here too, the market is going into multiple directions:

Home automation, connecting more sensors and devices in the home, some of them have cameras in them
Surveillance and security, where today it seems at least here in Israel, that fingerprint door locks are all the rage

Where does WebRTC play here? It might make things smoother to develop for the companies, but this doesn’t seem to be the case.

One thing that goes through all use cases above, is the existence of another solution – the video doorbell. Taken into buildings, this becomes an intercom system, which again – can make use of WebRTC. And why? Because it needs bidirectional support for audio at the very least, making WebRTC a suitable alternative.

Personal security

A totally different niche is the one of personal security.

This manifests itself as apps (and services) people can use to increase their security while going about in their daily tasks. Some of these apps connect you to friends and family while others to personal security agents. The WebRTC requirement here is the same for all cases – be able to conduct voice and video calls in real time.

Taken more broadly from the personal level, the same can be implemented in campuses, cities, events, etc.

Unique (?) challenges for WebRTC with camera hardware

There are some unique challenges for WebRTC when it comes to the surveillance space, and that’s mostly a matter of hardware.

Costs
- Hardware costs money. Not just the devices themselves, but their installation. This also means that hardware costs needs to be kept low in most of these systems, which means less processing power available on the cameras themselves or the xDR devices
- To drive costs down, CPUs won’t be as performant as the ones found in smartphones or PCs for example, and they would almost always rely heavily on hardware video encoding
Maintenance
- Many of these hardware systems come without subscription services. This means any firmware upgrades might or might not be available. It also means that such upgrades are sometimes clunky to get done on the devices, especially when they need to be handled remotely
- There’s a lot of physical maintenance as well involved. Cleaning lenses of cameras for example
Technology leaps
- You purchase a system. It has cameras and a xDR. Time passes. A couple of years. You decide you need more cameras, replace an existing one, whatever
- There’s improvements that took place. The system you have might not even be able to deal with the new cameras available today, and purchasing old ones might not be possible or economical anymore
- We had this when the system in our residential building broke. The DVR had a hard drive malfunction – it didn’t record anything anymore
  - It was impossible to replace, and buying a new old system wouldn’t be the right approach
  - Some of the cameras lost quality due to their analog coax cables (I was told this is an issue), and the predicament was we’d lose more of these cables in the coming 2-5 years anyways
  - So we had to shift the whole system to an IP based one. A technology leap
  - While I don’t foresee a move away from IP, I am sure many of these systems will change in the coming years in ways that would leave some of the old hardware unusable
Hybrid
- There are hybrid alternatives in this space. We ended up getting one for our building
- Due to the technology leaps, you end up with multiple types of sensors and cameras, from different generations and technologies
- The systems that cobbles it all together (the xDR in our case), can be one that manages them all
- Most installers won’t recommend it. It is mostly a necessary evil. Likely because it reduces the revenue of the installer and adds to the complexity of the installation and the system

Most of these issues won’t plague a software solution. But here, we end up in the real world simply because someone needs to go and install the physical cameras.

When figuring out the hardware platform to use, it is important to think of future trends and technology improvements that affect your implementation

In the case of surveillance, there’s WebRTC, future video codecs (AV1) and machine learning in the vision domain to think about. Probably also programmable photography that is bringing innovations to smartphones for a few years now

Ingress, egress and the concept of real time

Where to place WebRTC in the solution?

Since I write a lot about WebRTC, and this article is mostly about WebRTC in surveillance markets, it is THE biggest question to answer here.

There are two different places, and both are suitable, but not necessarily together in the same system.

Surveillance needs real time. Sometimes.

Egress

In our own residential building, I seldom care about the live feed from the cameras. It is to check if the front door to the building is open or not, or if there’s some area that got dirty (usually dog pee). Then most of the time is spent rewinding to figure out who caused the problem. Nothing here is considered real time in nature or requires sub second latency.

Elsewhere, real time might be critical on the viewer side (egress), which brings with it the question of whether WebRTC fits here well.

Ingress

Web cameras that directly stream out WebRTC to the world (or the xDR). Is that a benefit? What’s the value of it versus the existing camera technologies used?

I am not quite for or against this, as I am not really sure here. I’d say that a benefit here can be in the fact that it makes the whole technology stack simpler if you end up using WebRTC end-to-end instead of needing to switch protocols from the camera to the viewer. Just remember here that rewind and playback will likely require something other than WebRTC.

The main advantage of WebRTC here might be the removal of the need to transcode and translate across protocols and codecs. It makes xDR software simpler to write and reduces a lot of their CPU requirements, making the systems lighter and cheaper (the xDR – not the camera itself).

One more thing to think of is cameras that also require bidirectional audio. Because a security guard wants to announce or warn perpetrators, or because this is a video doorbell. There, WebRTC fits nicely, though again – not mandatory (I’d still try using it there more than elsewhere).

Going to introduce WebRTC to a surveillance system? Great. Check first where exactly within the whole architecture WebRTC fits and ask yourself why

Mobile or desktop?

Another important aspect of a surveillance system is where people go to watch the videos.

When we installed our own system, we were told that the mobile app is better than the PC app. In both, these were applications. But somehow for the consumers, it meant using the smartphone. It sucks. But yes – it sucks more on the desktop. Which is crazy, considering that what you’re trying to do is watch output coming from 4K cameras in order to identify people.

Then again, who is your customer?

If this is a large enterprise, where there’s going to be a fancy video wall of video feeds with a bored security guard looking at it, then should this be an application or would it be preferable to use a web application for it, with the help of WebRTC? It seems that much of the industry on the client side is looking for lightweight solutions that require less software installations, favoring browsers and… WebRTC.

And if you’re already doing WebRTC for one egress destination, you can use it for all others – browser and app based.

One more thing to consider – it is easier today to develop a web application than it is a native PC application. Cheaper and faster. Which means that supporting WebRTC if the desktop is your primary viewing device might be the right decision to make.

See if there’s a strong need for a zero-install or desktop viewing. This might well lead you towards WebRTC on the egress side

The age of Artificial Intelligence in surveillance tech

The biggest driver in this industry is machine learning and artificial intelligence. And not necessarily the Generative AI kind, but rather the kind that deals with object classification.

The challenge with surveillance is watching the damn cameras. You need eyeballs on screens. The good old motion detection removes a lot of noise (or more accurately, static), but it leaves much to be desired.

One of the elevators in my building, along with the video you get most hours of the day – empty. The bar at the bottom with the blue stripes marks when there’s actual movement.

Using machine learning, it will be easier to search for dogs, people, colors, items and other tidbits to figure out times of interest in the thousands of hours of boring videos, as well as act as “Google search” on recorded video feeds.

Doing all that in the cloud is possible, but expensive and tedious – how do you ship all the video, decode it, process it again, etc.

Doing it on the edge, on the device itself (the camera or the xDR) is preferable, but requires new hardware, so requires another technology leap and refresh.

WebRTC isn’t core for surveillance but it is critical

This is something to remember.

WebRTC isn’t core to surveillance. You don’t really need it to get surveillance cameras working, installed or connected to their xDR media servers. You don’t even need it to view videos – either “live” or as playback.

But, and that’s a big one – in some cases, having WebRTC is critical. Because your customer may want to be able to use web browsers and install nothing. He may want to be able to get bidirectional media. There might be a need to get video feeds that are at sub second latencies.

For these, WebRTC might not be a core competency, but they are critical to the successful delivery and deployment of your product. This translates into having a need to have that skill set in your team or be able to outsource it to someone with that skill set.

Where can I help, if at all?

Online WebRTC courses, to skill up engineers on this technology

Consulting, mostly around architecture decisions and technology stack selection Testing and monitoring WebRTC systems, via my role as Senior Director at Cyara (and the co-founder of testRTC)

The post Fitting WebRTC in the brave new world of webcams, security, surveillance and visual intelligence appeared first on BlogGeek.me.

Solving CPaaS vendor lock-in (as a customer and as a CPaaS vendor)

bloggeek - Tue, 09/12/2023 - 12:30

How to think and plan for CPaaS vendor lock-in when it comes to your WebRTC application implementation.

How can/should CPaaS vendors compete on winning customers? More than that, how can/should CPaaS vendors poach customers from other CPaaS vendors?

What prompted this article is the various techniques CPaaS vendors use and what they mean to customers – how should customers react to these techniques. I’ll focus on the Video API part of CPaaS – or to be more specific, the part that deals with WebRTC implementation.

Table of contents

What is CPaaS vendor lock-in?
The innovation in WebRTC that CPaaS is “killing”
CPaaS vendor poaching techniques and how to react to them
Build vs Buy – my first preference is ALWAYS buy (=CPaaS)

What is CPaaS vendor lock-in?

For me CPaaS (or Communication Platform as a Service) is a service that lets companies build their own communication experiences in a flexible manner. Usually done via APIs and requires developers, but recently, also via lowcode/nocode interactions (such as embedding an iframe).

A CPaaS vendor ends up defining its own interface of APIs which his customers are using to create these communication experiences.

That API interface is proprietary. There is no standard specification for how CPaaS APIs need to look or behave. This means that if you used such an API, and you want to switch to another CPaaS vendor – you’re going to need to do all that integration work all over again.

Think of it like switching from an Android phone to an iPhone or vice versa:

There’s a new interface you need to learn
- It might be similar since it practically used for doing the same things
- But it is also a bit “off”. The things you expect to be in one place are in another place
- The settings is done differently
- And the way you deal with the phone’s assistant (or Siri?) is different as well
You need to install all of your apps from scratch
- Find them in the app store, download them, install them
- Set them up by logging in
- Some of them you need to purchase separately all over again
- Others you won’t find… and you’ll need to look for alternative apps instead – or decide not to use that functionality any longer
The behavior will be different
- The background color of the apps
- They way you switch between screens is different
- The swipe “language” is also slightly different

In a way, you want the same experience (only better), but there’s going to be a learning curve and an adaptation curve where you familiarize yourself with the new CPaaS vendor and “make yourself at home”.

The vendor lock-in part is how much effort and risk will you need to invest and overcome in order to switch from one vendor to another – to call that other vendor your new home.

Vendor lock-in has 3 aspects to it in CPaaS:

Difference in the API interface. That’s a purely technical one. Low risk usually, with varying degree of effort
Behavioral differences. This has higher risk with unknown effort involved. While both CPaaS vendors do the “same” thing, they are doing it differently. And that difference is hidden behind how they behave. Your own application may rely on behavior that isn’t part of the standardized official interface and you will find out about it only once you test the migrated application on the new CPaaS vendor’s interface or later when things break in production
Integration differences. There are things outside the official interface you might have integrated with such as logs collection, understanding and handling error codes and edge cases, ETL processes, security mechanisms, etc. These things are the ones developers usually won’t account for when estimating the effort in the beginning and will likely be caught late in the migration process itself

Vendor lock-in is scary. Not because of the technical effort involved but because of the risks from the unknowns. The more years and the more interfaces, scenarios and code you have running on a CPaaS vendor, the higher the lock-in and risk of migration you are at.

The innovation in WebRTC that CPaaS is “killing”

Before WebRTC, we had other standards. RTP and RTCP came a lot before WebRTC.

We had RTMP, RTSP, SIP and H.323.

The main theme of all these standard specifications was that their focus has always been about standardizing what goes on over the network. They didn’t care or fret about the interface for the developer. The idea behind this was to enable using this standard on whatever hardware, operating system and programming language. Just read the spec and implement it anyway you like.

WebRTC changed all that (ignoring Flash here). We now have a specification where the API interface for the developer of a web application is also predefined.

WebRTC specifies what goes on the network, but also the JavaScript API in web browsers.

Here’s how I like explaining it in my slides:

One of the main advantages of WebRTC is that a developer who uses WebRTC in one project for one company can relatively easily switch to implement a different WebRTC project for another company. (that’s not really correct, but bear with me a little here)

We now could think of WebRTC just like other technologies – someone proficient in WebRTC is “comparable” to someone who worked with Node.js or SQL or other technologies. Whereas working with SIP or H.323 begs the question – which framework or implementation was used – learning a new one has its own learning curve.

Enter CPaaS…

And now the WebRTC API interface is no longer relevant. The CPaaS vendor’s SDK has its own interface indicating how things get done. And these may or may not bear any resemblance to the WebRTC API. Moreover – it might even try very hard to hide the WebRTC stack implementation from the developer.

This piece of innovation, where a developer using WebRTC can jump into new code of another project quickly is gone now. Because the interfaces of different CPaaS vendors aren’t standardized and don’t adhere to the standard WebRTC API interface (and they shouldn’t be – it isn’t because they are mean – it is because they offer a higher level of abstraction with more complex and complete functionality).

Not having the same interface across CPaaS vendors is one of the reasons we’ve started down this rabbit hole of exploring what CPaaS vendor lock-in is exactly.

CPaaS vendor poaching techniques and how to react to them

Every so often, you see one or more CPaaS vendors trying to grab a bit more market share in this space. Sometimes, it is about enticing customers who want to start using a CPaaS vendor. Other times it is focused on trying to poach customers from other CPaaS vendors.

When looking at the latter, here are the CPaaS vendor poaching techniques I’ve seen, how effective they are, and what you as a target company should think about them.

#1 – Feature list comparisons

The easiest technique to implement (and to review) is the feature list comparison.

In it, a CPaaS vendor would simply generate and share a comparison table of how its feature set is preferable over the popular alternatives.

For a company looking to switch, this would be a great place to start. You can skim through the feature list and see exactly what’s there in the platform you are currently using and the one you are thinking of switching to.

When looking at such a list, remember and ask yourself the following questions:

Is this list up to date? Oftentimes, these pages are created with big fanfare when a “poaching” or comparison project is initiated by the marketing department of a CPaaS vendor. But once done, it is seldom updated to reflect the latest versions (especially the latest version of the competitor). So take the comparison with a grain of salt. It is likely to be somewhat incorrect
Check what your experience is with the vendor you are using versus how it is reflected in the comparison table. Does the table describe things as you see them?
The features that look better “on paper” in this table for the vendor you plan on switching to. Do you need these features? Are they critical for you today or in the near future? Or are they just nice to have
The “greens” on the vendor making the comparison – are they on par with the other vendor or just a less comprehensive implementation of it? (for example, support for group calls – both vendors may support it, but one can get you to X users with open mics in a group call while the other can do 10X users)

I’ve had my fare share of reading, writing and responding to comparison tables. A long time ago (pre-WebRTC), we received inputs that our competitor can do almost 10 times the number of concurrent calls we are able to do with much higher throughput. Obviously, we created a task force to deal with it. The conclusion was simple – the competitor didn’t measure the network time at all – just CPU time in the machine. We weren’t measuring the same thing and his choice of metric meant he always looked better

Your role in this? To read between the lines and understand what wasn’t written. Always remember that this isn’t an objective comparison – it is highly skewed towards the author of it (otherwise, he wouldn’t be publishing it)

#2 – Performance comparisons

Here the intent of the CPaaS vendor is to show that his platform is superior in its performance. It can offer better quality, at lower bitrates and CPU use for larger groups.

If a vendor does it on his own, then potential customers will immediately view the results as suspect. This is why most of them use third party objective vendors to do these performance comparisons for them (at a cost).

We’ve done this at testRTC a couple of times – some publicly shared (for this one, I’ve placed my own reputation and testRTC’s reputation on the frontline, insisting not to name the other vendors) and others privately done. It is a fun project since it requires working towards a goal of figuring out how different CPaaS vendors behave in different scenarios.

Zoom did this as well, comparing itself to other CPaaS vendors. Agora answered in kind with a series of posts comparing themselves back to Zoom (where Zoom didn’t look as shiny).

Just remember a few things when reading such comparisons:

They were commissioned. They wouldn’t be published and shared if they weren’t showing what the CPaaS vendor wanted them to show
For me, it is more interesting to see how the setup of the performance tests was done and what was left out or missed in the comparison to begin with
- The types of machines and browsers selected
- Scenarios picked
- Reference applications used for each vendor
- How measurements are done
- Which metrics are selected for the comparison
Who the vendor was looking to compare himself to
The CPaaS vendor usually helps and tweaks his own platform to fit the scenarios selected, while the competing vendors have no say in which of their applications or samples are used and if or how they are optimized for the scenario (hint: they aren’t)

In the end, the fact that a CPaaS vendor performs better than another in a scenario you don’t need says nothing for you. Make sure to give more weight to the results of actual scenarios relevant to you, and be sure you understand what is really being compared

#3 – Guides, how-to’s and success stories

How do you make the migration of a customer from a different CPaaS vendor to your own? You write a migration document about it. A guide. Or a how-to. Or you get a testimonial or a success story from a customer willing to share publicly that he migrated and how life is so much better for him now.

These are mainly targeted at raising the confidence level for those who are contemplating switching, signaling them that the process isn’t risky and that others have taken this path successfully already.

As someone thinking of moving from one vendor to another, I’d seriously consider reaching out to the CPaaS vendor and ask the hard questions:

How easy is the migration really is
What challenges should one expect
Are there any common issues that migrating customers have bumped into
How many such customers do they have
Can they reach out and ask one of those who migrated to have a quick direct conversation with

Anecdotes and recipes are nice. What you are after is having more data points.

Read these guides and success stories. Try reading between the lines in them. Check if you have any open questions and then ask these questions directly. Gather as much information as you can to get a clearer picture

#4 – Reference applications

I wasn’t sure if this fits for migrating customers because it is a bit broader in nature. But here we are

In many cases, CPaaS vendors have reference applications available. Usually hosted on github. Just pull the code, compile, host and run it. You get an app that is “almost” ready for deployment.

You see how easy that was? Think how easy it is going to be to migrate to us with this great reference.

Remember a few things here:

Your workflow is likely different enough from the reference app that there’s work to be done here
In most cases, if you’ve built your application already on another vendor, using a reference app of another CPaaS vendor is close to impossible
Reference apps are just references. They usually don’t cover many of the edge cases that needs handling

From my point of view, reference apps are nice to get a taste of what’s possible and how the API of a CPaaS vendor gets used. But that’s about it. They are unlikely to be useful during the migration process itself

#5 – Shims and adaptors

They say imitation is the highest form of flattery. If that is true, then shims and adapters would fit well here.

In CPaaS, the most common one was supporting TwiML (that’s Twilio’s XML “language” for actions on telephony events). There’s also the idea/intent of having the whole API interface of another CPaaS vendor (or parts of it) supported directly by the poacher. The purpose of which is to make it easy to switch over.

Clearing things up a bit:

CPaaS vendor A has an API interface
CPaaS vendor B has a different API interface
To make it easier to switch from vendor A to vendor B, vendor B decides to create a piece of software that translates calls of A’s API interface to B’s API interface. This is usually called a shim or an adaptor

The result? If you’re using vendor A, theoretically, you can take the shim created by vendor B and magically without any investment, you migrate to vendor B. Problem solved

While this looks great on paper, I am afraid it has little chance of holding up in the real world . Here’s why:

The shim created is usually partial. Especially if vendor A offers a very rich interface (most vendors will, especially in the domain of video APIs and WebRTC)
Like reference applications, these shims don’t take good care of edge cases. Why? Because they aren’t used by many customers less customers = less investment
WebRTC is rather new, and CPaaS vendors have much to add, so every time vendor A updates his CPaaS and adds APIs to the interface – vendor B needs to invest in updating the shim. But is that even done once a shim is created? Or is it again, placed in the afterburner due the previous rule less customers = less investment
Behavior. Same API interface doesn’t necessarily mean the vendor’s platforms behave the same on the network. These changes are hard to catch… and might be even harder to resolve
Using a shim is nice, but if you want to use specific features available in vendor B’s interface – can you even do that if you’re doing everything via the shim? And is that the correct way to do things moving forward for you?

The thing is, that using a shim still means a ton of testing and headaches, but such that are hard to overcome.

If I had to switch between vendors, I’d ignore such shims altogether. For me they’re more of a trap than anything else.

Someone suggesting you use their shim for switching over to their CPaaS? Ignore them and just analyze what needs to be done as if there’s no shim available. You’ll thank me later

Build vs Buy – my first preference is ALWAYS buy (=CPaaS)

We’ve seen 5 different techniques CPaaS vendors use to try and poach customers from one another. For the most part, they are of the type of “buyers beware”. And yet, we do need to migrate from time to time from one CPaaS vendor to another. Market dynamics might force us to do so or just the need to switch to a better platform or offering.

Does that mean it would be best to go it alone and build your own platform instead of using a third party CPaaS vendor?

No.

Vendor lock-in isn’t necessarily a bad thing. My first preference is always to adopt a CPaaS vendor. And if not to adopt one, then to articulate very clearly why the decision to build is made.

What should you do when you start using a CPaaS vendor to make the transition to another vendor (or to your own platform) smoother in the distant future? Here are a few things to consider.

Limit the calls to the vendor’s API interface
- If you can make all of them from a single source file then great
- Even if not it is fine, but try not to call the vendor’s APIs and use their objects directly all over the place
- Having it all nicely compartmentalized will reduce the amount of changes needed during a migration
Consider building an abstraction layer
- While I hate this one, it appeals to some
- Create your own abstraction of the communications capabilities you need
- Have that abstraction a “standardized” internal interface you follow
- Implement the integration with the vendor as a class/object of that interface
- This enables you to implement the next vendor or your own platform as yet another class/object for the same interface in the future at some point.
- Risky. As this probably will require architectural and design changes once that time comes, but it might still be the decision that can get your company to move forward
Don’t use undocumented APIs and behaviors
- These will be harder to figure out in the future
- Making them harder to modify during a migration
Assume there’s no simple solution
- No silver bullet or magic solution here
- Which means that time invested in catering for future multiple vendors or seamless migration paths is time wasted
- Try to make the decisions here ones that don’t take more resources or time today due to some unknown future need – you are more likely to make a mistake in these decisions than you are to succeed in it

The post Solving CPaaS vendor lock-in (as a customer and as a CPaaS vendor) appeared first on BlogGeek.me.

WebRTC cracks the WHIP on OBS

webrtchacks - Tue, 08/22/2023 - 14:28

Open Broadcast Studio or OBS is an extremely popular open-source program used for streaming to broadcast platforms and for local recording. WebRTC is the open-source real time video communications stack built into every modern browser and used by billions for their regular video communications needs. Somehow these two have not formally intersected – that is […]

The post WebRTC cracks the WHIP on OBS appeared first on webrtcHacks.

WebRTC conferences – to mix or to route audio

bloggeek - Mon, 08/21/2023 - 12:30

How do you choose the right architecture for a WebRTC audio conferencing service?

Last month, Lorenzo Miniero published an update post on work he is doing on Janus to improve its AudioBridge plugin. It touched a point that I failed to write about for a long time (if at all), so I wanted to share my thoughts and views on it as well.

I’ll start with a quick explanation – Lorenzo is adding to Janus a lot of layers and flexibility that is needed by developers who are taking the route of mixing audio in WebRTC conferences. What I want to discuss here is when to use audio mixing and when not to use it. And as everything else, there usually isn’t a clear cut decision here.

Table of contents

What’s mixing and what’s routing in WebRTC?
Audio processing tools available for us in WebRTC
Mixing keeps the headaches away from the browser
Routing gets you better flexibility
Where the rubber hits the road – let’s talk use cases
Which will it be? MCU or SFU for your next audio meeting?

What’s mixing and what’s routing in WebRTC?

Group calls in WebRTC can take different shapes and sizes. For the most part, there are 3 dominant architectures for WebRTC multiparty calling: mesh, mixing and routing.

I’ll be focusing on mixing and routing here since they scale well to 100’s or more users.

Let’s start with the basics.

Assume there’s a conversation between 5 people. Each of these people can speak his mind and the others can hear him speaking. If all of these people are remote with each other and we now need to model it in WebRTC, we might think of it as something like this illustration:

This is known as a mesh network. Its biggest disadvantage for us (though there are others) is the messiness of it all – the number of connections between participants that grows polynomially with the number of users. The fact that we need to send out the same audio stream to all participants individually is another huge disadvantage. Usually, we assume (and for good reasons) that the network available to us is limited.

The immediate obvious solution is to get a central media server to mix all audio inputs, reducing all network traffic and processing from the users:

This media server is usually called an MCU (or a conferencing bridge). Users here “feel” as if they are in a session with only a single entity/user and the MCU is in charge of all the headaches on behalf of the users.

This mixer approach can be a wee bit expensive for the service provider and at times, not the most flexible of approaches. Which is why the SFU routed model was introduced, though mostly for video meetings. Here, we try to enjoy both worlds – we have the SFU route the media around, to try and keep bitrates and network use at reasonable levels while trying to reduce our hosting and media processing costs as service providers:

The SFU has become commonplace and the winning architecture model for video meetings almost everywhere. Voice only meetings though, have been somewhere in-between. Probably due to the existence and use of audio bridges a lot before WebRTC came to our lives.

This begs the question then, which architecture should we be using for our audio in group calls? Should we mix it in our media servers or just route it around like we do with video?

Before I go ahead to try and answer this question, there’s one more thing I’d like to go through, and that’s the set of media processing tools available to us today for audio in WebRTC.

Audio processing tools available for us in WebRTC

Encoding and decoding audio is the baseline thing. But other than that, there are quite a few media processing and network related algorithms that can assist applications in getting to the desired scale and quality of audio they need.

Before I list them, here are a few thoughts that came to mind when I collected them all:

This list is dynamic. It changes a bit every year or so, as new techniques are introduced
- An example for this is active speaker detection, appearing in a paper a decade ago
- First adopted by Jitsi
- Then added to mediasoup by the team at hopin
You can’t really use them all, all the time, for all use cases. You need to pick and choose the ones that are relevant to your use case, your users and the specific context you’re in
We now have a machine learning based tool as well. We will have more of these in a year or two for sure
It was a lot easier to compile this list now that we’ve finished recording and publishing all the lessons for the Higher-level WebRTC protocols course – we’ve covered most of these tools there in great detail

Audio level

There is an RTP header extension for audio level. This allows a WebRTC client to indicate what is the volume that can be found inside the encoded audio packet being sent.

The receiver can then use that information without decoding the packet at all.

What can one do with it?

Decide if you need to decode the packet at all or just discard it if there’s no or little voice activity or if the audio level is too low (no one’s going to hear what’s in there anyway).

You can replace it with DTX (see below) or not forward the packet in a Last-N architecture (see below).

Not mix its content with other audio channels (it doesn’t hold enough information to be useful to anyone).

DTX

Discontinuous transmission

If there’s nothing really to send – the person isn’t speaking but the microphone is open – then send “silence” but with less packets over the network.

That’s what DTX is about, and it is great.

In larger meetings, most people will listen and not speak over one another. So most audio streams will just be “silence” or muted. If they aren’t muted, then sending DTX instead of actual audio reduces the traffic generated. This can be a boon to SFUs who end up processing less packets.

An SFU media server can also decide to “replace” actual audio it receives from users (because they have a low audio level in them or because of Last-N decisions he is making) with DTX data when routing media around.

PLC

Packet Loss Concealment

Packets are going to be lost, but there would be content that still needs to be played back to the user.

You can decide to play silence, a repeat of the last heard packet, lower its volume a bit, etc.

This can be done both on the server side (especially in the case of an MCU mixer) or on the client side – where such algorithms are implemented in the browser already. SFUs can ignore this one, mostly since they don’t decode and process the actual media anyway.

At times, these can be done using machine learning, like Google’s proprietary WaveNetEq, which tries to estimate and predict what was in the missing packet based on past packets received.

Packet loss concealment isn’t great at all times, but it is a necessary evil.

RTX & NACK

Theoretically, you could use retransmissions for lost packets.

WebRTC does that mostly for video packets, but this can also find a home for audio.

It is/was a rather neglected area because PLC and Opus inband FEC techniques worked nicely.

For the time being, you’re likely to skip this tool, but it is one I’d keep an eye on if I were highly interested in audio quality advancements.

FEC and RED

Forward Error Correction is about sending redundant data that can be used to reconstruct lost packets. Redundancy coding is what we usually do for audio, which is duplicating encoded frames.

Audio bandwidth requirements are low, so duplicating frames doesn’t end up taxing much of our network, especially in a video call.

This approach enables us at a “low cost” to gain higher resiliency to packet losses.

This can be employed by the client sender, or even from the server side, beefing up what it received – both as an SFU or an MCU.

Check Philipp Hancke’s tal at Kranky Geek about Advanced in Audio Codecs

Then there’s the nuances and headaches of when to duplicate and how much, but that’s for another article.

Last-N

A known technicality in WebRTC’s implementation is that it only mixes the 3 loudest incoming audio channels before playing back the audio.

Why 3? Because 2 wasn’t enough and 4 seemed unnecessary is my guess. Also, the more sources you mix, the higher the noise levels are going to be, specially without good noise suppression (more on that below)

Well… Google just decided to remove that restriction. Based on the announcement, that’s because the audio decoding takes place in any case, so there’s not much of a performance optimization not to mix them all.

So now, you can decide if you want to mix everything (which you just couldn’t before) or if you want to mix or route only a few loudest volume (or most important) audio streams if that’s what you’re after. This reduces CPU and network load (depending on which architecture you are using).

Google Meet for example, is employing Last-3 technique, only sending up to 3 loudest audio streams to users in a meeting.

Oh, and if you want to dig deeper into the reasoning, there’s a nice Jitsi paper from 2016 explaining Last N.

Noise suppression: RNNoise and other machine learning algorithms

Noise suppression is all the rage these days.

RNNoise is a veteran among the ML-based noise suppression algorithms that is quite popular these days.

Janus for example, have added it to their AudioBridge and implemented optional RNNoise logic to handle channel-based noise suppression in their MCU mixer for each incoming stream.

Google added this in their Google Meet cloud – their SFU implementation passes the audio to dedicated servers that process this noise suppression – likely by decoding, noise suppression and encoding back the audio.

Many vendors today are introducing proprietary noise suppression to their solutions on the client side. These include Krisp, Dolby, Daily, Jitsi, Twilio and Agora – some via partnerships and others via self development.

Mixing keeps the headaches away from the browser

Why use an MCU for mixing your audio call? Because it takes all the implementation headaches and details away from the browser.

To understand some of what it entails on the server though, I’d refer you again to read Lorenzo’s post.

The great thing about this is that for the most part, adding more users means throwing more cloud hardware on the problem to solve it. At least up to a degree this can work well without thinking of scaling out, decentralization and other big words.

It is also how this was conducted for many years now.

Here are the tools I’d aim for in using for an audio MCU:

ToolUse?ReasoningAudio levelDecoding less streams will get higher performance density for the server. Use this with Last-N logicDTXBoth when decoding and while encodingPLCOn each incoming audio stream separatelyRTX & NACKTo early to do this todayFEC and REDToday, for an MCU, this would be rare to see as a supported featureConsider on outgoing audio streams; as well as enable for incoming streams from devicesLast-NLast-3 is a good default unless you have a specific user experience in mind (see below examples)Noise suppressionOn incoming channels, those that passed Last-N filtering, to clean them up before mixing the incoming streams together

Things to note with an audio MCU, is that the MCU needs to generate quite a few different outgoing streams. For 10 participants with 4 speakers (at Last-4 configuration), it would be something like this:

We have 5 separate mixers at play here:

1 mixing all 4 active speakers
4 mixing only 3 out of the 4 each time – we don’t want to send the person speaking his own audio mixed in the stream

Routing gets you better flexibility

Why do we use an SFU for audio conferences? Because we use it for video already… or because we believe this is the modern way of doing things these days.

When it comes to routing audio, the thing to remember is that we have a delicate balance between the SFU and the participants, each playing a part here to get a better experience at the end of the day.

Here are the tools I’d use for an audio SFU:

ToolUse?ReasoningAudio levelWe must have this thing implemented and enabled, especially since we really really really want to be able to conduct Last-N logic and not send each user all audio channels from all other participantsDTXWe can use this to detect silence as well here (and remove from Last-N logic). On the sending logic, the SFU can decide to DTX the channels in Last-N that are silent or at a low volume to save a bit of extra bandwidth (a minor optimization)PLCNot needed. We route the audio packets and let the participants fix any losses that take placeRTX & NACKTo early to do this todayFEC and REDThis can be added on the receiver and sender side in the SFU to improve audio quality. Adding logic to dynamically device when and how much redundancy based on network conditions is also an advantage hereLast-NLast-3 is a good default. Probably best to keep this at most at Last-5 since the decision here means more CPU use on the participants’ sideNoise suppressionNot needed. This can be done on the participants’ side

In many ways, an audio SFU is simpler to implement than an audio MCU, but tweaking it just right to gain all the benefits and optimizations from the client implementation is the tricky part.

Where the rubber hits the road – let’s talk use cases

As with everything else I deal with, which approach to use depends on the circumstances. One of the main deciding criteria in this case is going to be the use case you are dealing with and the scenario you are solving this for.

Here are a few that came to mind.

Gateway to the old world

The first one is borderline “obvious”.

Before WebRTC, no one really did an audio conference using an SFU architecture. And if they did, it was unique, proprietary and special. The world revolved and still revolves around MCU and mixing audio bridges.

If your service needs to connect to legacy telephony services, existing deployments of VoIP services running over SIP (or god forbid H.323), connect to a large XMPP network – whatever it may be – that “other” world is going to be running as an MCU. Each device is likely capable of handling only one incoming audio stream.

So trying to connect a few users from your service (no matter if you are using an SFU or an MCU) would need to mix these users when connecting them to the legacy service.

Video meetings with mixed audio

There are services that decide to use an SFU to route video streams and an MCU for the audio streams.

Sometimes, it is because the main service started as an audio service (so an audio bridge was/is at the heart of the service already) and video was bolted on the platform. Sometimes it is because gatewaying to the old world is central to the service and its mindset.

Other times, it is due to an effort to reduce the number of audio streams being sent around, or to reduce the technical requirements of audio only participants.

Whatever the reason, this is something you might bump into.

The big downside of such an approach is the loss of lip synchronization. There is no practical way you can synchronize a single audio stream that represents mixed content of multiple video streams. In fact, no lip synchronization with any of the video streams takes place…

Usually, the excuse I’ll be hearing is that the latency difference isn’t noticeable and no one complained. Which begs the question – why do we bother with lip synchronization mechanisms at all then? (we do because it does matter and is noticeable – especially when the network is slightly bumpier than usual)

Experience the crowd

Think of a soccer game. 50,000 people in a stadium. Rawring when there’s a goal or a miss.

With Last-3 audio streams mixed, you wouldn’t be hearing anything interesting when this takes place “remotely” for the viewers.

The same applies to a virtual online concert.

Part of the experience you are trying to convey is the crowds and the noises and voices they generate.

If we’re all busy reducing noise levels, suppressing it, picking and choosing the 2-3 voices in the crowd to mix, then we just degrade the experience.

Crowds matter in some scenarios. And keeping their experience properly cannot be done by routing audio streams around. Especially not when we’re starting to talk about hundreds of more active participants.

This case necessitates the use of MCU audio bridging. And likely a distributed approach the moment the numbers of users climb higher.

Metaverse and spatial audio

The metaverse is coming. Or will be. Maybe. Now that Apple Vision Pro is upon us. But even before that, we’ve seen some metaverse use cases.

One thing that comes to mind here is the immersion part of it, which leads to spatial audio. The intent of hearing multiple sounds coming from different directions – based on where the speaker is.

This means several things:

For each user, the angle and distance (=volume level) of each other person speaking is going to be different
That Last-3 strategy doesn’t work anymore. If you can distinguish directionness and volume levels individually, then more sources might need to be “mixed” here

Do you do that on the client side by way of an SFU implementation, or would it be preferable to do this in an MCU implementation?

And what about trying to run concerts in the metaverse? How do you give the notion of the crowds on the audio side?

These are questions that definitely don’t have a single answer.

In all likelihood, in some metaverse cases, the SFU model will be the best architectural approach while in others an MCU would work better.

Recording it all

Not exactly a use case in its own right, but rather a feature that is needed a lot.

When we need to record a session, how do we go about doing that?

Today, in at least 99% of the time that would be by mixing all audio and video sources and creating a single stream that can be played as a “regular” mp4 file (or similar).

Recording as a single stream means using an MCU-like solution. Sometimes by implementing it in a headless browser (as if this is a silent participant in the session) and other times by way of dedicated media servers. The result is similar – mixing the multiple incoming streams into a single outgoing one that goes directly to storage.

The downside of this, besides needing to spend energy on mixing something that people might never see (which is a decision point to which architecture to pick for example), is that you get to view and hear only a single viewpoint of a single user – since the mixed recording is already “opinionated” based on what viewpoint it took.

We can theoretically “record” the streams separately and then play them back separately, but that’s not that simple to achieve, and for the most part, it isn’t commonplace.

A kind of a compromise we see today with professional recording and podcast services is to record by mixed and separated audio streams. This allows post production to take either based on the mixing needs, but done manually.

Which will it be? MCU or SFU for your next audio meeting?

We start with this, and we will end with this.

It depends.

You need to understand your requirements and from there see if the solution you need will be based on an MCU, and SFU or both. And if you need help with figuring that out, that’s what my WebRTC courses are for – check them out.

The post WebRTC conferences – to mix or to route audio appeared first on BlogGeek.me.

10 Years of webrtcHacks – merch and stats

webrtchacks - Mon, 07/24/2023 - 22:11

webrtcHacks celebrates our 10th birthday today 🎂. To commemorate this day, I’ll cover 2 topics here: Our new merch store Some stats and trends looking back on 10 years of posts We have the Merch In the early days of webrtcHacks, co-founder Reid Stidolph ordered a bunch of stickers which proved to be extremely popular. […]

The post 10 Years of webrtcHacks – merch and stats appeared first on webrtcHacks.

WebCodecs, WebTransport, and the Future of WebRTC

webrtchacks - Tue, 07/18/2023 - 14:30

Explore the future of Real-Time Communications with WebrtcHacks as we delve into the use of WebCodecs and WebTransport as alternatives to WebRTC's RTCPeerConnection. This comprehensive blog post features interviews with industry experts, a review of potential WebCodecs+WebTransport architecture, and a discussion on real-time media processing challenges. We also examine performance measurements, hardware encoder issues, and the practicality of these new technologies.

The post WebCodecs, WebTransport, and the Future of WebRTC appeared first on webrtcHacks.

New: Higher-Level WebRTC Protocols course

bloggeek - Mon, 07/17/2023 - 12:30

A new Higher-level WebRTC protocols course and discounts, available for a limited period of time.

Over a year ago, Philipp Hancke came to me with the idea of creating a new set of courses. Ones that will dig deeper into the heart of the protocols used in WebRTC. This being a huge undertaking, we decided to split it into several courses, and focus on the first one – Low-level WebRTC protocols.

We received positive feedback about it, so we ended up working on our second course in this series – Higher-level WebRTC protocols.

Why the need for additional WebRTC courses?

There is always something more to learn.

The initial courses at WebRTC Course were focused on giving an understanding of the different components of WebRTC itself and on getting developers to be able to design and then implement their application.

What was missing in all that was a closer look at the protocols themselves. Of looking at what goes on in the network, and being able to understand what goes over the wire. Which is why we started off with the protocols courses.

Where the Low-level WebRTC protocols looks at directly what goes to the network with WebRTC, our newer Higher-level WebRTC protocols is taking it up one level:

This time, we’re looking at the protocols that make use of RTP and RTCP to make the job of real time communications manageable.

If you don’t know exactly what header extensions are, and how they work (and why), or the types of bandwidth estimation algorithms that WebRTC uses – and again – how and why – then this course is for you.

If you know RTP and RTCP really well, because you’ve worked in the video conferencing industry, or have done SIP for years – then this course is definitely for you.

Just understanding the types of RTP header extensions that WebRTC ends up using, many of them proprietary, is going to be quite a surprise for you.

Our WebRTC Protocols courses

Got a use case where you need to render remote machines using WebRTC? These require sitting at the cutting edge of WebRTC, or more accurately and a slightly skewed angle versus what the general population does with WebRTC (including Google).

Taking upon yourself such a use case means you’ll need to rely more heavily on your own expertise and understanding of WebRTC.

There are now 2 available protocols courses for you:

Low-level WebRTC protocols
Higher-level WebRTC protocols (half-complete. Call it a work in progress)

And there are 2 different ways to purchase them:

Each one separately – low and high
As part of the bigger ALL INCLUDED WebRTC Developer bundle (the Higher-level course was just added to it)

You should probably hurry though…

There’s a 40% discount on the Higher-level WebRTC protocols course. This early-bird discount will be available until the end of this month ($180 instead of $300)
There’s also a 20% discount on all courses and ebooks. Call it a summer sale – this one is available using discount code SUMMER

Check out my WebRTC courses

The post New: Higher-Level WebRTC Protocols course appeared first on BlogGeek.me.

Cloud gaming, virtual desktops and WebRTC

bloggeek - Mon, 07/03/2023 - 13:30

WebRTC is an important technology for cloud gaming and virtual desktop type use cases. Here are the reasons and the challenges associated with it.

Google launched and shut down Stadia. A cloud gaming platform. It used WebRTC (yay), but it didn’t quite fit into Google’s future it seems.

That said, it does shed a light on a use case that I’ve been “neglecting” in my writing here, though it was and is definitely top of mind in discussions with vendors and developers.

What I want to put in writing this time is cloud gaming as a concept, and then alongside it, all virtual desktops and cloud rendering use cases.

Let’s dig in

Table of contents

The rise and (predictable?) fall of Google Stadia
WebRTC, Cloud gaming and the challenges of real time
What’s in cloud gaming anyway?
Beyond cloud gaming – virtual desktop, remote desktop and cloud rendering
Why use WebRTC in cloud gaming
Enter our WebRTC Protocols courses

The rise and (predictable?) fall of Google Stadia

Google Stadia started life as Project Stream inside Google.

Technically, it made perfect sense. But at least in hindsight, the business plan wasn’t really there. Google is far remote from gaming, game developers and gamers.

On the technical side, the intent was to run high end games on cloud machines that would render the game and then have someone play the game “remotely”. The user gets a live video rendering of the game and sends back console signals. This meant games could be as complex as they need be and get their compute power from cloud servers, while keeping the user’s device at the same spec no matter the game.

Source: Google

I’ve added the WebRTC text on the diagram from Google – WebRTC was called upon so that the player could use a modern browser to play the game. No installation needed. This can work nicely even on iOS devices, where Apple is adamant about their part of the revenue sharing on anything that goes through the app store.

Stadia wanted to solve quite a few technological challenges:

Running high end console games on cloud machines
Remotely serving these games in real time
Playing the game inside a browser (or an equivalent)

And likely quite a few other challenges as well (scaling this whole thing and figuring out how to obtain and keep so many GPUs for example).

Technically, Stadia was a success. Businesswise… well… it shut down a little over 3 years after its launch – so not so much.

What Stadia did though, was show that this is most definitely possible.

WebRTC, Cloud gaming and the challenges of real time

To get cloud gaming right, Google had to do a few things with WebRTC. Things they haven’t really needed too much when the main thing for WebRTC at Google was Google Meet. These were lowering the latency, dealing with a larger color space and aiming for 4K resolution at 60 fps. What they got virtually for “free” with WebRTC was its data channel – the means to send game controller signals quickly from the player to the gaming machine in the cloud.

Lets see what it meant to add the other three things:

4K resolution at 60 fps

Google aimed for high end games, which meant higher resolutions and frame rates.

WebRTC is/was great for video conferencing resolutions. VGA, 720p and even 1080p. 4K was another jump up that scale. It requires more CPU and more bandwidth.

Luckily, for cloud gaming, the browser only needs to decode the video and not encode it. Which meant the real issue, besides making sure the browser can actually decode 4K resolutions efficiently, was to conduct efficient bandwidth estimation.

As an algorithm, bandwidth estimation is finely tuned and optimized for given scenarios. 4K and cloud gaming being a new scenario, meant that bitrates that were needed weren’t 2mbps or even 4mbps but rather more in the range of 10-35mbps.

The built-in bandwidth estimator in WebRTC can’t handle this… but the one Google built for the Stadia servers can. On the technical side, this was made possible by Google relying on sender-side bandwidth estimation techniques using transport-cc.

Lower latency: playout delay

Remember this diagram?

It can be found in my article titled With media delivery, you can optimize for quality or latency. Not both.

WebRTC is designed and built for lower latency, but in the sub-second latency, how would you sort the latency requirements of these 3 activities?

Nailing a SpaceX rocket landing
Playing a first shooter game (as old as I am, that means Doom or Quake for me)
Having an online meeting with a potential customer

WebRTC’s main focus over the years has been online meetings. This means having 100 milliseconds or 200 milliseconds delay would be just fine.

With an online game? 100 milliseconds is the difference between winning and losing.

So Google tried to reduce latency even further with WebRTC by adding a concept of Playout Delay. The intent here is to let WebRTC know that the application and use case prefers playing out the media earlier and sacrificing even further in quality, versus waiting a bit for the benefit of maybe getting better quality.

Larger color space

Video conferencing and talking heads doesn’t need much. If you recall, with video compression what we’re after is to lose as much as we can out of the original video signal and then compress. The idea here is that whatever the eye won’t notice – we can make do without.

Apparently, for talking heads we can lose more of the “color” and still be happy versus doing something similar for an online game.

To make a point, if you’ve watched Game of Thrones at home, then you may remember the botch they had with the last season with some of the episodes that ended up being too dark for television. That was due to compression done by service providers…

So far this is my favorite screenshot from #BattleForWinterfell #GameofThrones pic.twitter.com/6uI45SjPG7

— Lady Emily (@GreatCheshire) April 29, 2019

While different from the color space issue here, it goes to show that how you treat color in video encoding matters. And it differs from one scenario to another.

When it comes to games, a different treatment of color space was needed. Specifically, moving from SDR to HDR, adding an RTP header extension in the process to express that additional information.

–

Oh, and if you want to learn more about these changes (especially resolution and color space), then make sure to watch this Kranky Geek session by YouTube about the changes they had to make to support Stadia:

What’s in cloud gaming anyway?

Here’s the thing. Google Stadia is one end of the spectrum in gaming and in cloud gaming.

Throughout the years, I’ve seen quite a few other reasons and market targets for cloud gaming.

Types of cloud games

Here are the ones that come out of the top of my head:

High end gaming. That’s the Google Stadia use case. Play a high end game anywhere you want on any kind of device. This reduces the reliance and need to upgrade your gaming hardware all the time
- You’ll find NVIDIA, Amazon Luna and Microsoft xCloud focused in this domain
- How popular/profitable this is is still questionable
Console gaming. PlayStation, Xbox, Switch. Whatever. Picking a game and playing without waiting to download and install is great. It also allows reducing/removing the hard drive from these devices (or shrinking them in size)
Mobile games. You can now sample mobile apps and games before downloading them, running them in the cloud. Other things here? You could play games of other users using their account and the levels they reached instead of slaving your way there
Retro/emulated games. There’s a large and growing body of games that can’t be played on today’s machines because the hardware for them is too old. These can be emulated, and now also played remotely as cloud games. How about playing a PlayStation 2 game? Or an old and classing SEGA arcade game? Me thinking Golden Axe

Improved gameplay

Why not even play these games with others remotely?

My son recently had a sit down with 4 other friends, all playing on Xbox together a TMNT game. It was great having them all over, but you could do it remotely as well. If the game doesn’t offer remote players, by pushing it to the cloud you can get that feature simply because all users immediately become remote players.

At this stage, you can even add a voice conference or a video call to the game between the players. Just to give them the level of collaboration they can get out of playing the likes of Fortnite. Granted, this requires more than just game rendering in the cloud, but it is possible and I do see it happen with some of the vendors in this space.

Beyond cloud gaming – virtual desktop, remote desktop and cloud rendering

Lower latencies. Bigger color space. Higher resolutions. Rendering in the cloud and consuming remotely.

All these aren’t specific to cloud gaming. They can easily be extended to virtual desktop and remote desktop scenarios.

You have a machine in the cloud – big or small or even a cluster. That “machine” handles computations and ends up rendering the result to a virtual display. You then grab that display and send it to a remote user.

One use case can just be a remote desktop a-la VNC. Here we’re actually trying to get connected from one machine to another, usually in a private and secure peer-to-peer fashion, which is different from what I am aiming for here.

Another, less talked about is doing things like Photoshop operations in the cloud. For the poor sad people like me who don’t have the latest Mac Pro with the shiny M2 Ultra chip, I might just want to “rent” the compute power online for my image or video editing jobs.

I might want to open a rendered 3D view of a sports car I’d like to buy, directly from the browser, having the ability to move my view around the car.

Or it might just be a simple VDI scenario, where the company (usually a large financial institute, but not only) would like the employees to work on Chromebook machines but have nothing installed or stored in them – all consumed by accessing the actual machine and data in their own corporate data center or secure cloud environment.

A good friend of mine asked me what PC to buy for himself. He needed it for work. He is a lawyer. My answer was the lowest end machine you can find would do the job. That saved him quite a lot of money I am guessing, and he wouldn’t even notice the difference for what he needs it for.

But what if he needs a bit more juice and power every once in a while? Can renting that in the cloud make a difference?

What about the need to use specialized software that is hard to install and configure? Or that requires a lot of collaboration on large amounts of data that need to be shared across the collaborators?

Taking the notion and capabilities of cloud gaming and applying them to non-gaming use cases can help us with multiple other requirements:

CPU and memory requirements that can’t be met with a local machine easily
The need to maintain privacy and corporate data in work from home environments
Zero install environment, lowering maintenance costs

Do these have to happen with WebRTC? No

Can they happen with WebRTC? Yes

Would changing from proprietary VDI environments to open standard WebRTC in browsers improve things? Probably

Why use WebRTC in cloud gaming

Why even use WebRTC for cloud gaming or more general cloud rendering then?

With cloud gaming, we’re fine doing it from inside a dedicated app. So WebRTC isn’t really necessary. Or is it?

In one of our recent WebRTC Insights issues we’ve highlighted that Amazon Luna is dropping the dedicated apps in favor of the web (=WebRTC). From that article:

“We saw customers were spending significantly more time playing games on Luna using their web browsers than on native PC and Mac apps. When we see customers love something, we double down. We optimized the web browser experience with the full features and capabilities offered in Luna’s native desktop apps so customers now have the same exact Luna experience when using Luna on their web browsers.”

Browsers are still a popular enough alternative for many users. Are these your users too?

If you need or want web browser access for a cloud gaming / cloud rendering application, then WebRTC is the way to go. It is a slightly different opinion than the one I had with the future of live streaming, where I stated the opposite:

“The reason WebRTC is used at the moment is because it was the only game in town. Soon that will change with the adoption of solutions based on WebTransport+WebCodecs+WebAssembly where an alternative to WebRTC for live streaming in browsers will introduce itself.”

Why the difference? It is all about the latency we are willing to accommodate:

Your mileage may vary when it comes to the specific latency you’re aiming for, but in general – live streaming can live with slightly higher latency than our online meetings. So something other than WebRTC can cater for that better – we can fine tune and tweak it more.

Cloud gaming needs even lower latency than WebRTC. And WebRTC can accommodate for that. Using something else that is unproven yet (and suffers from performance and latency issues a bit at the moment) is the wrong approach. At least today.

Enter our WebRTC Protocols courses

Taking upon yourself such a use case means you’ll need to rely more heavily on your own expertise and understanding of WebRTC.

Over a year ago I launched with Philipp Hancke the Low-level WebRTC Protocols course. We’re now recording our next course – Higher-level WebRTC Protocols.

If you are interested in learning more about this, be sure to join our waiting list for once we launch the course

Join the course waiting list

Oh, and I’d like to thank Midjourney for releasing version 5.2 – awesome images

The post Cloud gaming, virtual desktops and WebRTC appeared first on BlogGeek.me.

Apple Vision, VR/AR, the metaverse and what it means to the web and WebRTC

bloggeek - Mon, 06/19/2023 - 13:30

The Apple Vision pro is a new VR/AR headset. Here are my thoughts on if and how it will affect the metaverse and WebRTC.

There were quite a few interesting announcements and advances made in recent months that got me thinking about this whole area of the metaverse, augmented reality and virtual reality. All of which culminated with Apple’s unveiling last week of the Apple Vision Pro. For me, the prism from which I analyze things is the one of communication technologies, and predominantly WebRTC.

A quick disclaimer: I have no clue about what the future holds here or how it affects WebRTC. The whole purpose of this article is for me to try and sort my own thoughts by putting them “down on paper”.

Let’s get started then

Table of contents

The Apple Vision Pro
Video calling, FaceTime, televisions and AR
- FaceTime and Apple TV
- FaceTime and Apple Vision Pro
Lifelike video meetings: Just like being there
Volumetric video coding
Is the metaverse web based?
Where would one fit WebRTC in the metaverse, AR and VR?

The Apple Vision Pro

Apple just announced its Vision Pro VR/AR headset. If you’re reading this blog, then you know about this already, so there isn’t much to say about it.

For me? This is the first time that I had this nagging feeling for a few seconds that I just might want to go and purchase an Apple product.

Most articles I’ve read were raving about this – especially the ones who got a few minutes to play with it at Apple’s headquarters.

AR/VR headsets thus far have been taking one of the two approaches:

AR headsets were more akin to “glasses” that had an overhead display on them which is where the augmentation took place with additional information being displayed on top of reality. Think Google Glass
VR headsets which you wear a whole new world on top of your head, looking at a video screen that replaces the real world altogether

Apple took the middle ground – their headset is a VR headset since it replaces what you see with two high resolution displays – one for each eye. But it acts as an AR headset – because it uses external cameras on the headset to project the world on these displays.t

The end result? Expensive, but probably with better utility than any other alternative, especially once you couple it with Apple’s software.

Video calling, FaceTime, televisions and AR

Almost at the sidelines of all the talks and discussions around Apple Vision Pro and the new Mac machines, there have been a few announcements around things that interest me the most – video calling.

FaceTime and Apple TV

One of the challenges of video calling has been to put it on the television. This used to be called a lean back experience for video calling, in a world predominantly focused on lean forward when it comes to video calling. I remember working on such proof of concepts and product demos with customers ~15 years ago or more.

These never caught on.

The main reason was somewhere between the cost of the hardware, maintaining privacy with a livingroom camera and microphone positioning/noise.

By tethering the iPhone to the television, the cost of hardware along with maintaining privacy gets solved. The microphones are now a lot better than they used to – mostly due to better software.

Apple, being Apple, can offer a unique experience because they own and control the hardware – both of the phone and the set-top box. Something that is hard for other vendors to pull off.

There’s a nice concept video on the Apple press release for this, which reminded me of this Facebook (now Meta) Portal presentation from Kranky Geek:

Can Android devices pull the same thing, connected to Chromecast enabled devices maybe? Or is that too much to ask?

Do television and/or set-top box vendors put an effort into a similar solution? Should they be worried in any way?

Where could/should WebRTC play a role in such solutions, if at all?

FaceTime and Apple Vision Pro

How do you manage video calls with a clunky AR/VR headset plastered on your face?

First off, there’s no external camera “watching you”, unless you add one. And then there’s the nagging thing of… well… the headset:

Apple has this “figured out” by way of generating a realistic avatar of you in a meeting. What is interesting to note here, is that in the Apple Vision Pro announcement video itself, Apple made a three important omissions:

They don’t show how the other people in the meeting see the person with the Vision headset on
There’s only a single person with a Vision headset on, and we have his worldview, so again, we can’t see how others with a Vision headset look like in such a call
How do you maintain eye contact, or even know where the user is looking at? (a problem today’s video calling solutions have as well)

What do the people at the meeting see of her? Do they see her looking at them, or the side of her head? Do they see the context of her real-life surroundings or a virtual background?

I couldn’t find any person who played with the Apple Vision Pro headset and reported using FaceTime, so I am assuming this one is still a work in progress. It will be really interesting to see what they come up with once this is released to market, and how real life use looks and feels like.

Lifelike video meetings: Just like being there

Then there’s telepresence. This amorphous thing which for me translates into: “expensive video conferencing meeting rooms no one can purchase unless they are too rich”.

Or if I am a wee bit less sarcastic – it is where we strive to with video conferencing – what would be the ultimate experience of “just like being there” done remotely if we had the best technology money can buy today.

Google Project Starline is the current poster child of this telepresence technology.

The current iteration of telepresence strives to provide 3D lifelike experience (with eye contact obviously). To do so while maintaining hardware costs down and fitting more environments and hardware devices, it will rely on AI – like everything else these days.

The result as I understand it?

Background removal/replacement
Understanding depth, to be able to generate a 3D representation of a speaker on demand and fit it to what the viewer needs, as opposed to what the cameras directly capture

Now look at what FaceTime on an Apple Vision Pro really means:

Generate a hyper realistic avatar representation of the person – this sounds really similar to removing the background and using cameras to generate a 3D representation of the speaker (just with a bit more work and a bit less accuracy).

Both Vision Pro and Starline strive for lifelike experiences between remote people. Starline goes for a meeting room experience, capturing the essence of the real world. Vision Pro goes after a mix between augmented and virtual reality here – can’t really say this is augmented, but can’t say this is virtual either.

A telepresence system may end up selling a million units a year (a gross exaggeration on my part as to the size of the market, if you take the most optimistic outcome), whereas a headset will end up selling in the tens of millions or more once it is successful (and this is probably a realistic estimate).

What both of these ends of the same continuum of a video meeting experience do is they add the notion of 3D, which in video is referred to as volumetric video (we need to use big fancy words to show off our smarts).

And yes, that does lead me to the next topic I’d like to cover – volumetric video encoding.

Volumetric video coding

We have the metaverse now. Virtual reality. Augmented reality. The works.

How do we communicate on top of it? What does a video look like now?

The obvious answer today would be “it’s a 3D video”. And now we need to be able to compress it and send it over the network – just like any other 2D video.

The Alliance for Open Media, who has been behind the publication and promotion of the AV1 video codec, just published a call for proposals related to volumetric video compression. From the proposal, I want to focus on the following tidbits:

The Alliance’s Volumetric Visual Media (VVM) Working Group formed in February 2022 this is rather new
It is led by Co-Chairs Khaled Mammou, Principal Engineer at Apple, and Shan Liu, Distinguished Scientist and General Manager at Tencent Apple… me thinking Vision Pro
The purpose is the “development of new tools for the compression of volumetric visual media” better compression tools for 3D video

This being promoted now, on the same week Apple Vision Pro comes out might be a coincidence. Or it might not.

The founding members include all the relevant vendors interested in AR/VR that you’d assume:

Apple – obviously
Cisco – WebEx and telepresence
Google – think Project Starline
Intel & NVIDIA – selling CPUs and GPUs to all of us
Meta – and their metaverse
Microsoft – with Teams, Hololens and metaverse aspirations

The rest also have vested interest in the metaverse, so this all boils down to this:

AR/VR requires new video coding techniques to enable better and more efficient communications in 3D (among other things)

Apple Vision Pro isn’t alone in this, but likely the one taking the first bold steps

The big question for me is this – will Apple go off with its own volumetric video codecs here, touting how open they are (think FaceTime open) or will they embrace the Alliance of Open Media work that they themselves are co-chairing?

And if they do go for the open standard here, will they also make it available for other developers to use? Me thinking… WebRTC

Is the metaverse web based?

Before tackling the notion of WebRTC into the metaverse, there’s one more prerequisite – that’s the web itself.

Would we be accessing the metaverse via a web browser, or a similar construct?

For an open metaverse, this would be something we’d like to have – the ability to have our own identity(ies) in the metaverse go with us wherever we go – between Facebook, to Roblox, through Fortnite or whatever other “domain” we go to.

Last week also got us this sideline announcement from Matrix: Introducing Third Room TP2: The Creator Update

Matrix, an open source and open standard for decentralized communications, have been working on Third Room, which for me is a kind of a metaverse infrastructure for the web. Like everything related to the metaverse, this is mostly a work in progress.

I’d love the metaverse itself to be web based and open, but it seems most vendors would rather have it limited to their own closed gardens (Apple and Meta certainly would love it that way. So would many others). I definitely see how open standards might end up being used in the metaverse (like the work the Alliance of Open Media is doing), but the vendors who will adopt these open standards will end up deciding how open to make their implementations – and will the web be the place to do it all or not.

Where would one fit WebRTC in the metaverse, AR and VR?

Maybe. Maybe not.

The unbundling of WebRTC makes it both an option while taking us farther away from having WebRTC as part of the future metaverse.

Not having the web means no real reliance on WebRTC.

Having the tooling in WebRTC to assist developers in the metaverse means there’s incentive to use and adopt it even without the web browser angle of it.

WebRTC will need at some point to deal with some new technical requirements to properly support metaverse use cases:

Volumetric video coding
Improve its spatial audio capability
The number of audio streams that can be mixed (3 are the max today)

We’re still far away from that target, and there will be a lot of other technologies that will need to be crammed in alongside WebRTC itself to make this whole thing happen.

Apple’s new Vision Pro might accelerate that trajectory of WebRTC – or it might just do the opposite – solidify the world of the metaverse inside native apps.

—

I want to finish this off with this short piece by Jason Fried: The visions of the future

It looks at AR/VR and generative AI, and how they are two exact opposites in many ways.

Recently I also covered ChatGPT and WebRTC – you might want to take a look at that while at it.

The post Apple Vision, VR/AR, the metaverse and what it means to the web and WebRTC appeared first on BlogGeek.me.

Livestream this Friday: WebCodecs, WebTransport, and the Future of WebRTC

webrtchacks - Tue, 06/06/2023 - 14:10

Here at webrtcHacks we are always exploring what’s next in the world of Real Time Communications. One area we have touched on a few times is the use of WebCodecs with WebTransport as an alternative to WebRTC’s RTCPeerConnection. There have been several recent experiments by Bernard Aboba – WebRTC & WebTransport Co-Chair and webrtcHacks regular, […]

The post Livestream this Friday: WebCodecs, WebTransport, and the Future of WebRTC appeared first on webrtcHacks.

Is WebRTC really free? The costs of running a WebRTC application

bloggeek - Mon, 06/05/2023 - 12:30

Is WebRTC really free? It is open source and widely used due to it. But it isn’t free when it comes to running and hosting your own WebRTC applications.

If you are new to WebRTC, then start here – What is WebRTC?

Time to answer this nagging question:

Is WebRTC really free?

One of the reasons that WebRTC is the most widely used developer technology for real time communications in the world is that it is open source. It helps a lot that it comes embedded and available in all modern browsers. That means that anyone can use WebRTC for any purpose they see fit, without paying any upfront licensing fee or later on royalties. This has enabled thousands of companies to develop and launch their own applications.

But does that mean every web application built with WebRTC is free? No. WebRTC may well be free, but whatever is bolted on top of it might not be. And then there are still costs involved with getting a web application online and dealing with traffic costs.

For that reason, in this article, I’ll be touching on why WebRTC really is free, and what you have to factor in for it if you want to get your own WebRTC application.

Table of contents

Yes. WebRTC itself is completely free
However, there are costs involved with running a WebRTC application
1. The cost of developing a WebRTC application
2. How expensive it is to optimize a WebRTC implementation
3. Hosting and maintenance costs of a WebRTC application
- Cloud costs
- Keeping the lights costs
Build vs buy a WebRTC infrastructure
So, how free is WebRTC, really?

Yes. WebRTC itself is completely free

Since I am sure you didn’t really go read that other article – I’ll suggest it here again: What is WebRTC?

The TL;DR version of it?

The WebRTC software library is open sourced under a permissible open source license. That means its source code is available to everyone AND that individuals and companies can modify and use it anywhere they wish without needing to contribute back their changes. It makes it easier for commercial software to be developed with it (even when no changes or improvements are made to the base WebRTC library – just because of how corporate lawyers are).

You see? WebRTC really is free.

Google “owns” and maintains the main WebRTC library implementation. Everyone benefits from this. That siad, they aren’t doing this only from the goodness of their heart – they have their own uses for WebRTC they focus on.

However, there are costs involved with running a WebRTC application

While you don’t have to pay anything for WebRTC itself, there’s the application you develop, publish and then maintain. There are costs that come into play here – and considerable ones. These costs can vary depending on your requirements.

I’d like to split the costs here into 3 components:

The cost of developing a WebRTC application
How expensive it is to optimize a WebRTC implementation
Hosting and maintenance costs of a WebRTC application

1. The cost of developing a WebRTC application

The first thing you can put as a cost is to build the WebRTC application itself.

Here, as in all other areas, there’s more demand than supply when it comes to skilled WebRTC engineers. So much so that I had to write an article about hiring WebRTC developers – and I still send this link multiple times a month when asked about this.

Here too, you should split the cost into two parts:

How much does it cost to develop your application?
The WebRTC part of the application – how much investment do you need to put on it?

Since everything done in WebRTC requires skilled engineers (that are scarce when it comes to WebRTC expertise), you can safely assume it is going to be a wee bit more expensive than you estimate it to be.

2. How expensive it is to optimize a WebRTC implementation

I know what you’re going to say. Your WebRTC application is going to be awesome. Glorious. Superb. It is going to be so good that it will wipe the floor with the existing solutions such as Zoom, Google Meet and Microsoft Teams.

That kind of a mentality is healthy in an entrepreneur, but a dose of reality is necessary here:

You can’t out-do Google in quality with WebRTC
- At least not if you’re going to butt head to head
- Remember that they’re the ones who maintain WebRTC and implement it inside Chrome
- And if you have the skillset to actually deliver on this one, it means you don’t need to read this article…
These vendors have large teams
- Larger than what you are going to put out there
- Almost definitely larger than what you’re going to budget for in the next 3 years
- In man-years they are going to out-class you on pure media quality
- Especially when the focus is on improving it in our industry at the moment
- These vendors also need to deal with how Google runs WebRTC in the browser

This brings me to the need to optimize what you’re doing on an ongoing basis.

Ever since the pandemic, we’ve seen a growing effort in the leading vendors in this space to improve and optimize quality. This manifests itself in the research they publish as well as features they bring to the market. Here are a few examples:

Larger meeting sizes
Lower CPU use
Newer audio and video codecs
Introduction of AI algorithms to the media pipeline

You should plan for ongoing optimization of your own as well. Your customers are going to expect you to keep up with the industry. The notion of “good enough” works well here, but the bar of what is “good enough” is rising all the time.

Such optimizations are also needed not only to improve quality, but also to reduce costs.

Factor these costs in…

3. Hosting and maintenance costs of a WebRTC application

I had a meeting the other day. A founder of a startup who had to use WebRTC because customers needed something live and interactive. That component wasn’t at the core of his application, but not having it meant lost deals and revenue. It was a mandatory capability needed for a specific feature.

He complained about WebRTC being expensive to operate. Mainly because of bandwidth costs.

We can split WebRTC maintenance costs here into two categories: cloud costs, keeping the lights on costs.

Cloud costs

That startup founder was focused on cloud costs.

When we look at the infrastructure costs of web applications, there’s the usual CPU, memory, storage and network. We might be paying these directly, or indirectly via other managed and serverless services.

With WebRTC, the network component is the biggest hurt. Especially for video applications. You can reduce these costs by going to 2nd tier IaaS vendors or by hosting in “no-name” local data centers, but if you are like most vendors, you’re likely to end up on Amazon, Microsoft or Google cloud. And there, bandwidth costs for outgoing traffic are high.

WebRTC is peer to peer, but:

Not all sessions can go peer to peer. Some must be relayed via TURN servers
Large group calls in most cases will mean going through the cloud with your bandwidth to WebRTC media servers
All commercial WebRTC services I know have server components that gobble up bandwidth

And the more successful you become – the more bandwidth you’ll consume – and the higher your cloud costs are going to be.

You will need to factor this in when developing your application, especially deciding when to start optimizing for costs and bandwidth use.

Keeping the lights costs

Then there’s the “keeping the lights” costs.

WebRTC changes all the time. Things get deprecated and removed. Features change behavior over time. New features are added. You continually need to test that your application does not break in the upcoming Chrome release. Who is going to take care of all that in your WebRTC application?

You will also need to understand the way your WebRTC application is used. Are users happy? Are there areas you need to invest in with further optimization? Observability (=monitoring) is key here.

Keeping the lights on has its own set of costs associated with it.

Build vs buy a WebRTC infrastructure

Buying your WebRTC infrastructure by using managed services like CPaaS vendors is expensive. But then again, building your own (along with optimizing and maintaining it) is also expensive.

Roughly speaking, this is the kind of a decision table you’ll see in front of you:

BuildBuyPros Customized to your specific need
Ownership of the solution and ability to modify with changing requirements
Better control over costs Short time to market
Low initial cost
Less of a need for a highly skilled team of WebRTC expertsCons Time consuming. Longer time to market
High initial development costs
Ongoing maintenance costs
Finding/sourcing skilled WebRTC developers Cost at scale can be an issue
Harder to differentiate on the media layer

There’s also a middleground, where you can source/buy certain pieces and build others. Here are a few examples/suggestions:

Consider paying for a managed TURN service while building your own WebRTC application
Signaling can be outsourced using the likes of PubNub, Pusher and Ably
You can get your testing and monitoring needs from testRTC (a company I co-founded)

You can also start with a CPaaS vendor and once you scale and grow, invest the time and money needed to build your own infrastructure – once you’ve proven your application and got to product-market-fit.

So, how free is WebRTC, really?

Part of WebRTC’s claim to fame is its nature as an open source and thus free software for building interactive web applications. While the technology itself is indeed free of charge and offers numerous freedoms, there are still costs associated with running a WebRTC application.

When we had to launch our own video conferencing service some 25 years ago, we had to put an investment of several millions of dollars along with an engineering team for a period of a couple of years. Only to get to the implementation of a media engine.

WebRTC gives that to you for “free”. And it is also kind enough to be pre-integrated in all modern browsers.

What Google did with WebRTC was to reduce the barrier of entry to real time communication drastically.

Creating a WebRTC application isn’t free – not really. But it does come with a lot of alternatives that bring with them freedom and flexibility.

The post Is WebRTC really free? The costs of running a WebRTC application appeared first on BlogGeek.me.

WebRTC media resilience: the role FEC, RED, PLC, RTX and other acronyms play

bloggeek - Mon, 05/22/2023 - 13:00

How WebRTC media resilience works – what FEC, RED, PLC, RTX are and why they are needed to improve media quality in real-time communications.

Networks are finicky in nature, and media codecs even more so.

With networks, not everything sent is received on the other end, which means we have one more thing to deal with and care about when it comes to handling WebRTC media. Luckily for us, there are quite a few built-in tools that are available to us. But which one should we use at each point and what benefits do they bring?

This is what I’ll be focusing on in this article.

Table of contents

Networks are lossy
Media codecs are lossy (and sensitive)
Types of WebRTC media correction
PLC: Packet Loss Concealment
- Audio PLC
- Video PLC
RTX: Retransmissions
RED: REDundancy Encoding
FEC: Forward Error Correction
Making sense of WebRTC media correction

Networks are lossy

Communication networks are lossy in nature. This means that if you send a packet through a network – there’s no guarantee of that packet reaching the other side. There’s also no guarantee that packets are reached in the order you’ve sent them or in a timely fashion, but that’s for another article.

This is why almost everything you do over the internet has this nice retransmission mechanism tucked away somewhere deep inside as an assumption. That retransmission mechanism is part of how TCP works – and for that matter, almost every other transport protocol implemented inside browsers.

The assumption here is that if something is lost, you simply send it again and you’re done. It may take a wee bit longer for the receiver to receive it, but it will get there. And if it doesn’t, we can simply announce that connection as severed and closed.

We call and measure that “something is lost” aspect of networks as packet loss.

Stripping away that automatic assumption that networks are reliable and everything you send over them is received on the other side is the first important step in understanding WebRTC but also in understanding real-time transport protocols and their underlying concepts.

Media codecs are lossy (and sensitive)

Media codecs are also lossy but in a different way. When an audio codec or a video codec needs to encode (=compress) the raw input from a microphone or a camera, what they do is strip the data out of things they deem unnecessary. These things are levels of perceived quality of the original media.

I remember many years ago, sitting at the dorms in the university and talking about albums and CDs. One of the roommates there was an audiophile. He always explained how vinyl albums have better audio quality than CDs and how MP3 just ruins audio quality. Me? I never heard the difference.

Perceived quality might be different between different people. The better the codec implementation, the more people will not notice degraded quality.

Back to codecs.

Most media codecs are lossy in nature. There are a few lossless ones, but these are rarely used for real time communications and not used in WebRTC at all. The reason we use lossy codecs is to have better compression rates:

Taking 1080p (Full HD) video at 30 frames per second will result in roughly 1.5Gbps of data. Without compressing it – it just won’t work. We’re trying to squeeze a lot of raw data over networks, and as always, we need to balance our needs with the resources available to us.

To compress more, we need:

To reduce what we care about to the its bare minimum (the lossy aspect of the codecs we use)
More CPU and memory to perform the compression
Make every bit we end up with matter

That last one is where media codecs become really sensitive.

If every bit matters, then losing a bit matters. And if losing a bit matters, then losing a whole packet matters even more.

Since networks are bound to lose packets, we’re going to need to deal with media packets missing and our system (in the decoder or elsewhere) needing to fill that gap somehow

ChatGPT meets WebRTC: What Generative AI means to Real Time Communications

bloggeek - Mon, 05/08/2023 - 13:00

ChatGPT is changing computing and as an extension how we interact with machines. Here’s how it is going to affect WebRTC.

ChatGPT became the service with the highest growth rate of any internet application, reaching 100 million active users within the first two months of its existence. A few are using it daily. Others are experimenting with it. Many have heard about it. All of us will be affected by it in one way or another.

I’ve been trying to figure out what exactly does a “ChatGPT WebRTC” duo means – or in other words – what does ChatGPT means for those of us working with and on WebRTC.

Here are my thoughts so far.

Table of contents

Crash course on ChatGPT
- BI, AI and Generative AI
- The stellar rise of ChatGPT
Why ChatGPT and WebRTC are like oil and water
What have people done with ChatGPT and WebRTC so far?
Broadening the scope: Generative AI
Fitting Generative AI to the world of RTC
Is the future of WebRTC generative (AI)?

Crash course on ChatGPT

Let’s start with a quick look at what ChatGPT really is (in layman terms, with a lot of hand waving, and probably more than a few mistakes along the way).

BI, AI and Generative AI

I’ll start with a few slides I cobbled up for a presentation I did for a group of friends who wanted to understand this.

ChatGPT is a product/service that makes use of machine learning. Machine learning is something that has been marketed a lot as AI – Artificial Intelligence. If you look at how this field has evolved, it would be something like the below:

We started with simple statistics – take a few numbers, sum them up, divide by their count and you get an average. You complicate that a bit with weighted average. Add a bit more statistics on top of it, collect more data points and cobble up a nice BI (Business Intelligence) system.

At some point, we started looking at deep learning:

Here, we train a model by using a lot of data points, to a point that the model can infer things about new data given to it. Things like “do you see a dog in this picture?” or “what is the text being said in this audio recording?”.

Here, a lot of 3 letter acronyms are used like HMM, ANN, CNN, RNN, GNN…

What deep learning did in the past decade or two was enable machines to describe things – be able to identify objects in images and videos, convert speech to text, etc.

It made it the ultimate classifier, improving the way we search and catalog things.

And then came a new field of solutions in the form of Generative AI. Here, machine learning is used to generate new data, as opposed to classifying existing data:

Here what we’re doing is creating a random input vector, pushing it into a generator model. The generator model creates a sample for us – something that *should* result in the type of thing we want created (say a picture of a dog). That sample that was generated is then passed to the “traditional” inference model that checks if this is indeed what we wanted to generate. If it isn’t, we iteratively try to fine tune it until we get to a result that is “real”.

This is time consuming and resource intensive – but it works rather well for many use cases (like some of the images on this site’s articles that are now generated with the help of Midjourney).

So…

We started with averages and statistics
Moved to “deep learning”, which is just hard for us to explain how the algorithms got to the results they did (it isn’t based on simple rules any longer)
And we then got to a point where AI generates new data

The stellar rise of ChatGPT

The thing is that all this thing I just explained wouldn’t be interesting without ChatGPT – a service that came to our lives only recently, becoming the hottest thing out there:

The Most Important Chart In 100 Years https://t.co/Ypcsqi0AWJ #AI #GPT #ChatGPT #technology @JohnNosta pic.twitter.com/QjMroVZ7cG

— Kyle Hailey (@kylelf_) February 16, 2023

ChatGPT is based on LLMs – Large Language Models – and it is fast becoming the hottest thing around. No other service grew as fast as ChatGPT, which is why every business in the world now is trying to figure out if and how ChatGPT will fit into their world and services.

Why ChatGPT and WebRTC are like oil and water

So it begged the question: what can you do with ChatGPT and WebRTC?

Problem is, ChatGPT and WebRTC are like oil and water – they don’t mix that well.

ChatGPT generates data whereas WebRTC enables people to communicate with each other. The “generation” part in WebRTC is taken care of by the humans that interact mostly with each other on it.

On one hand, this makes ChatGPT kinda useless for WebRTC – or at least not that obvious to use for it.

But on the other hand, if someone succeeds to crack this one up properly – he will have an innovative and unique thing.

What have people done with ChatGPT and WebRTC so far?

It is interesting to see what people and companies have done with ChatGPT and WebRTC in the last couple of months. Here are a few things that I’ve noticed:

Arin Sime decided to ask ChatGPT about the future of WebRTC. Nice, but not really something that gets WebRTC and ChatGPT more integrated with one another
LiveKit shows how to connect ChatGPT to a live WebRTC video call. The result is mindbogglingly good – practically giving voice to ChatGPT
Twilio showcases a similar thing – connecting ChatGPT to their Programmable Voice service. Slightly less compelling but just as practical
Then there’s the whole transcription space, where you see ChatGPT and its ilk used for the generation of summaries and action items from the meeting transcription

In LiveKit’s and Twilio’s examples, the concept is to use the audio source from humans as part of prompts for ChatGPT after converting them using Speech to Text and then converting the ChatGPT response using Text to Speech and pass it back to the humans in the conversation.

Broadening the scope: Generative AI

ChatGPT is one of many generative AI services. Its focus is on text. Other generative AI solutions deal with images or sound or video or practically any other data that needs to be generated.

I have been using MidJourney for the past several months to help me with the creation of many images in this blog.

Today it seems that in any field where new data or information needs to be created, a generative AI algorithm can be a good place to investigate. And in marketing-speak – AI is overused and a new overhyped term was needed to explain what innovation and cutting edge is – so the word “generative” was added to AI for that purpose.

Fitting Generative AI to the world of RTC

How does one go about connecting generative AI technologies with communications then? The answer to this question isn’t an obvious or simple one. From what I’ve seen, there are 3 main areas where you can make use of generative AI with WebRTC (or just RTC):

Conversations and bots
Media compression
Media processing

Here’s what it means

Conversations and bots

In this area, we either have a conversation with a bot or have a bot “eavesdrop” on a conversation.

The LiveKit and Twilio examples earlier are about striking a conversation with a bot – much like how you’d use ChatGPT’s prompts.

A bot eavesdropping to a conversation can offer assistance throughout a meeting or after the meeting –

It can try to capture to essence of a session, turning it into a summary
Help with note taking and writing down action items
Figure out additional resources to share during the conversation – such as knowledge base items that reflect what a customer is complaining about to a call center agent

As I stated above, this has little to do with WebRTC itself – it takes place elsewhere in the pipeline; and to me, this is mostly an application capability.

Media compression

An interesting domain where AI is starting to be investigated and used is media compression. I’ve written about Lyra, Google’s AI enabled speech codec in the past. Lyra makes assumptions on how human speech sounds and behaves in order to send less data over the network (effectively compressing it) and letting the receiving end figure out and fill out the gaps using machine learning. Can this approach be seen as a case of generative AI? Maybe

Would investigating such approaches where the speakers are known to better compress their audio and even video makes sense?

How about the whole super resolution angle? Where you send video at resolutions of WVGA or 720p and then having the decoder scale them up to 1080p or 4K, losing little in the process. We’re generating data out of thin air, though probably not in the “classic” sense of generative AI.

I’d also argue that if you know the initial raw content was generated using generative AI, there might be a better way in which the data can be compressed and sent at lower bitrates. Is that something worth pursuing or investigating? I don’t know.

Media processing

Similar to how we can have AI based codecs such as Lyra, we can also use AI algorithms to improve quality – better packet loss concealment that learns the speech patterns in real time and then mimics them when there’s packet loss. This is what Google is doing with their WaveNetEQ, something I mentioned in my WebRTC unbundling article from 2020.

Here again, the main question is how much of this is generative AI versus simply AI – and does that even matter?

Is the future of WebRTC generative (AI)?

ChatGTP and other generative AI services are growing and evolving rapidly. While WebRTC isn’t directly linked to this trend, it certainly is affected by it:

Applications will need to figure out how (and why) to incorporate generative AI with WebRTC as part of what they offer
Algorithms and codecs in WebRTC are evolving with the assistance of AI (generative or otherwise)

Like any other person and business out there, you too should see if and how does generative AI affects your own plans.

The post ChatGPT meets WebRTC: What Generative AI means to Real Time Communications appeared first on BlogGeek.me.

News from Industry

The Hidden AV1 Gift in Google Meet

Twilio exits video APIs, further focusing on voice, SMS and Segment

Third time’s a charm: WebRTC Insights, 3 years in

Qotom Q20321G9 fanless PC

Zooming in on remote education and WebRTC

WebRTC in telehealth: More than just HIPAA compliance

No. I am not ok

Fitting WebRTC in the brave new world of webcams, security, surveillance and visual intelligence

Solving CPaaS vendor lock-in (as a customer and as a CPaaS vendor)

WebRTC cracks the WHIP on OBS

WebRTC conferences – to mix or to route audio

10 Years of webrtcHacks – merch and stats

WebCodecs, WebTransport, and the Future of WebRTC

New: Higher-Level WebRTC Protocols course

Cloud gaming, virtual desktops and WebRTC

Apple Vision, VR/AR, the metaverse and what it means to the web and WebRTC

Livestream this Friday: WebCodecs, WebTransport, and the Future of WebRTC

Is WebRTC really free? The costs of running a WebRTC application

WebRTC media resilience: the role FEC, RED, PLC, RTX and other acronyms play

ChatGPT meets WebRTC: What Generative AI means to Real Time Communications

Pages

Using the greatness of Parallax

Yet more available pages

Responsive grid

Typography

About

WITH A RICH FOOTER

Recent comments

Main menu

News from Industry

Pages

Using the greatness of Parallax

Yet more available pages

Responsive grid

Typography

Main menu

User login