Well, this is awkward: Telos Alliance on TV audio technology and live cloud production 12 months on

By Martin Dyster, Telos Alliance VP business development.

At this time of year, I generally take a look back at my predictions from 12 months ago and see what I got right, where I was a little wide of the mark and what missed the target completely. So, let’s just say I may have been a bit over-optimistic back in December 2023 when it comes to gauging the likely progress in the development of virtual audio products for live cloud production (LCP)!

My three predictions for 2023 were as follows:

• That we’d see the launch of a range of virtualised audio mixers from multiple brands that can scale well enough to handle most sizes of production from mono right up to Atmos.
• There would be Intercom solutions available from established brands that enabled a hybrid approach to mix and match software and hardware panels.
• These products would be able to interconnect with autonomy to create virtualized audio and communications subsystems that can be deployed quickly and with ease.

Looking at my first prediction, there has been some very exciting progress made from one particular brand. The folks at Audiotonix have been very busy indeed and have played their cards early with the announcement of a joint venture between the otherwise competing brands of SSL and Calrec. The New Heights (a most appropriate name) DSP mixing core will enable users the choice of using either SSL T-Series or Calrec Argo hardware consoles to remotely control the new virtual engine, with ground to cloud and cloud to cloud signal flows handled by either Dante Connect and/or NDI.

It’s unusual for big brands to break ranks and unveil their development strategies so early during a product’s lifecycle (full release isn’t due until mid-2024), but by doing so, Audiotonix has staked its claim and become the market leader in a field that is sure to grow exponentially over the next five to 10 years. The lessons it will learn from proof of concept projects with customers looking to move into LCP will be invaluable in the years to come. It is a bold but admirable move.

Whispers and rumours

Virtual Intercom has kept me very busy since early 2021. I can say with some pride that the work we’ve done at Telos to further the cause of communications in LCP has shaken up the industry and put Telos Alliance on the TV map. I’m not writing this piece to promote a brand, so I’ll simply invite the reader to check it out for themselves. However, like the majority of manufacturers in the mixing console market, the Intercom industry has not embraced virtualisation and continues to develop products that are architected around a central processing core, either a monolithic matrix or a DSP engine that performs the same job, but more efficiently.

We work in a very small industry and there are constant whispers and rumours about new products on the horizon, but for now it seems that the marquee brands of the intercom world are content with what they have. It leaves a space on the fringes for businesses like Telos, AEQ, GreenGo and Unity to flourish as we build products for those customers who are looking for something different. From my experience, we find ourselves often integrating our own solutions with those more established brands to create a hybrid architecture that serves the market well.

My third prediction looked at the tricky issue of interconnecting subsystems in the virtual space. I mentioned Dante Connect and NDI a couple of paragraphs back, and both are technologies that ease the pain. When I talk about interconnectivity, I am describing the ability to trunk audio across WAN or VPN between the on premises systems (at event control rooms, pitch side microphones, commentary booths, IFBs etc.) to the cloud and back.

Most of these signals must be delivered with minimal latency, synchronously; for contribution they will mainly be uncompressed, but in the case of communications they can sometimes be asynchronous and also compressed to save data costs. Dante Connect covers the requirement for synchronous uncompressed signal transport, and NDI can handle both formats since it offers the user choice. Of course, NDI was developed primarily for video workflows, but it supports up to 256 channels of synchronous audio per video signal with both advertisement and discovery inbuilt. With the new custodians of the format, Vizrt, actively supporting NDI audio-only solutions and companies like Lama, Avsono, and Sienna developing dedicated audio signal management products for NDI users, NDI looks set to complement or even rival Dante as a go-to problem solver for virtual audio.

Other established means of audio transport also exist with WebRTC (using the OPUS codec) as the preferred solution for our own Telos communications platform and many other voice-based applications. WebRTC benefits from onboard encryption, Forward Error Correction (FEC), and variable bitrate.

SRT is another video transport format that supports audio and AWS also has tools designed to simplify interconnectivity between cloud instances and on prem subsystems. AWS Transit Gateway and Direct Connect are designed to interconnect Amazon VPCs and on premises networks (Transit Gateway), and between AWS Data Centres to span across hybrid facilities (Direct Connect).

On the surface, it looks like my third prediction from last year has been somewhat fulfilled, but although there is choice, what is lacking is something that is standards-based. There’s an implied relationship between virtualized audio platforms and audio-over-IP that exists because the infrastructure on either side of the on prem to/from cloud transport codecs is AoIP based. Dante supports AES67 / SMPTE ST2110-30 of course, but Dante Connect currently does not, and NDI and SRT are proprietary. Multicast AES67 / ST2110-30 cannot easily or reliably traverse between virtual instances without tools like AWS Direct Connect or Transit Gateway.

Looking forward to 2024 and for the third year in a row, I’ll make a prediction. Aside from the emergence of new and exciting virtual audio production technology, the complex mechanisms to connect those that exist will evolve and the landscape become clearer. It has to happen, not just for audio but for video and control so that LCP can flourish for the cost saving and efficiency reasons that make it so appealing to many broadcasters.