Live Sports Production: Audio Acquisition
Achieving the deeply engaging entertainment experience sports fans demand starts all the way back at the source, with a high quality microphone and a highly skilled audio engineer.
All 12 articles in this series are now available in our free eBook ‘Live Sports Production’ – download it HERE.
All articles are also available individually:
While the challenges around transport, timing, delay, communication and interoperability are not insignificant, not to mention ensuring that personnel are where they need to be when they need to be there, the heart of audio acquisition remains unchanged. The concept of garbage in, garbage out is so prevalent that it even has its own acronym (GIGO, obviously), and in this sense its principals haven’t really changed since Alexander Graeme Bell and Emil Berliner introduced the first microphone technologies in the 1870s. Incoming signals are still only as good as the microphone capturing them, and if the microphone is not set correctly, you can’t just fix it with EQ.
So it’s unsurprising that NEP Australia’s Michael Saliba is adamant that everything he does starts with the microphone.
The Microphone Is Still The Source
Michael Saliba, NEP Australia: “After the preamp, everything is evolving at pace, but it all starts at the microphone and our clients still demand the best sound quality at every stage of the signal. They want microphone home runs to the preamp and they don’t want cables going through multiple joins. The good news is that these things are definitely getting better, and getting the preamps closer to the source is much easier than it used to be, especially with stageboxes linked over fiber.”
“We are always looking to improve the quality of the audio from the source, and it is important to work with our suppliers to get exactly what we and our clients need. Although a lot of the more immersive capture we do uses Schoeps ORTF-3D equipment, all of our shotgun mics and RF kit are built around Sennheiser; we have a really good working relationship with them and we often work together to get specific products over the line, right down to the cables that we want to spec.”
Hybrid Workflows
Beyond the preamp, that’s a different story. From an audio acquisition perspective, how does NEP Australia’s workflow differ between working on site in a truck and working remotely? Is there such a clear distinction or are all modern audio workflows hybrid?
Michael Saliba, NEP Australia: “It’s always a hybrid. Since the outset of NEP Australia’s Hub model, our philosophy has always been built on the concept of locating anyone anywhere, so our approach has always been crew-based; the first question is to ascertain where the crews are available. They might be in Sydney, but we might have a brilliant Audio Director (AD) in Brisbane, and just because we built the hub doesn’t mean we have to use it. It’s always a combination.”
“We’ve covered cricket matches where we’ve had people capturing audio on the stumps in the truck, and the production mixing back at the Hub. A big part of how we make that possible is that the desks are mirrored rather than mixing in parallel, so if the mix engineer is doing stumps, those stump faders will move in the hub, and if either engineer needs to go for a break, the other can take over on the same fader.”
Achieving Consistency
In our interview about the Audio Control Room (ACR), Saliba talked about consistency and how all its ACRs use the same kit to not just ensure the same results, but also to make the environments very familiar. But that consistency doesn’t end there.
“We want the OB facilities to be able to cope with whatever we need to do, so we never want to change the configuration of the trucks if we can help it. For example, we don’t dial up all 64 bidirectional channels on all our stageboxes; we break them down to more granular numbers so we can access as many or as few preamps as we need for a specific job.”
“A good example of that is live golf. Our bigger OB facilities each have five stageboxes which are all capable of 32 in and 16 out; in golf we might use more than 20 stageboxes on a single production, but we might only dial in multiples of eight channels in each box. Operating at that level of granularity helps us manage connectivity because we’ve already standardized and managed our stream counts.”
“These levels of scalability and flexibility also influence the equipment we purchase, so we’re always looking at equipment that’s easily upgradeable by adding more resources to the network. We’ve already made sure that we have an adequate cabling infrastructure to deal with change, so if we have to move things around, we can easily do it.”
“And the same goes for how we build our showfiles; although a lot of our trucks are capable of doing independent productions, we don’t have standalone files for the truck, they are all built in a Hub file. That way, any job can turn into a Hub job because it has all of the resources already there.”
Simplicity is key to making complex networks operate flawlessly, while standardizing setups and never overcommitting resources means that once the network is up and running, NEP Australia has very little engineering work outside of path switching through its TFC broadcast controller platform. And according to Saliba, it has also had an effect on how NEP Australia’s key audio staff approach their jobs.
“It has definitely all had an effect on how our ADs work because it removes some of the customizability that ADs are able to bring to the production. In the past, as long as the AD put the right things on the right outputs, what happened in the middle didn’t matter so much. That’s no longer the case, but pushing for a standard across the board is worth it as it makes everyone’s life a lot easier.”
Connectivity
Connectivity is another vital part of the puzzle, and with dozens of venues across Australia connected to its two Hubs, it is the Telstra Distributed Production Network (DPN) circuit that does a lot of the heavy lifting. We asked Saliba about how NEP Australia designed the infrastructure between its Hubs, trucks and onsite sources, as well as how it guarantees bandwidth and uptime.
“We try to give everyone the same experience wherever the mix position is, and we’ve designed the infrastructure to be non-blocking so it isn’t limiting in any way.”
“We use the Telstra DPN across our network. When Telstra launched its DPN circuit it operated at 50 Gb redundant connectivity at bigger venues, and we operate bigger circuits of about 400 Gb capacity between our Sydney and Melbourne Hubs. Our aim was always to provide the highest bandwidth and lowest latency connectivity before we have places like Perth and New Zealand on the grid. Our broadcast orchestration platform controls the switching of all those feeds and the stadiums are already set up so that the flows can be routed with just a couple of button presses.”
“The network uses ST 2022-7 redundant protection switching and diverse paths so it deals in purple network management switches with a red and blue media plane. When we describe five 10 Gb lanes at a 50 Gb venue, there are five primary lanes of red connectivity and five lanes of secondary blue multicast paths, which means we can lose one whole path and still be in control. Actually, part of our testing procedure for all connected OBs is to shut an entire path down, so we’ll shut a whole red path down and be held up on blue. That’s true of our control as well, which is the purple part of the network and ensures that control traffic can be routed through the primary or secondary network.”
“From an audio perspective, there is no requirement for us to stick to the traditional 64 channels anymore in SMPTE, but we do that anyway. We connect four 64 channel flows at all times between the truck and the OB, and we can increase it up to six for bigger productions. This covers all of the audio and intercoms, and whatever else is necessary.”
“In the future we are looking at bigger capacity circuits. It’s still very easy to oversubscribe a 10 Gb lane if we push the boundaries of the production. Moving ahead, we’re looking at introducing more 400 Gb circuits. This doesn’t have much of an effect on the audio – we’re already operating uncompressed, and we need far less bandwidth – but it will have a bigger effect on what we do with visual content.”
Aligning Time
Despite taking advantage of Telstra’s low-latency connectivity and mirroring consoles to take advantage of edge processing where possible, audio signals still have to travel and Australia is – famously – the sixth largest country on the planet. That takes time, and it is why latency and delay is such a big deal with remote production workflows like NEP Australia’s Hub infrastructure. On some broadcasts, NEP is not only managing delay but is dealing with up to eight different timing planes.
“Most of the sports we cover have a standard formula and we have designations for things like talent, ball effects and stadium FX, with automated delay for cameras to keep things lip synced. We can deal with a lot of different vision perspectives, so we might have a POV in a commentary box, or a camera shooting a referee that’s miles away or some other kind of specialty camera.”
“It’s not up to the vision guys to time align all of that with the audio, but it is the audio department that is responsible if it is not in sync. The best way to get that right is to trigger things off the vision mixer and dial in delays where appropriate for that source, which means we run slip delays for a lot of the stuff that we do.”
Looking Ahead
The Hub model has fundamentally changed the way NEP Australia works, but the company is not complacent.
“We’re going through transition periods that are less than five years now, and the big change that I can already see is moving into cloud compute based software defined infrastructure where services are license-based. We used to spec a truck to be overly capable to service the biggest production; I’m already designing systems on the smallest possible license with a view to upgrade it for the time period I need it. It’s a very different way of looking at things because the audio core hardware is the same, whether it’s licensed or not; it’s all capable.”
“We are already decentralizing our processing with the Hub model, but with licensing you can do it in a more affordable way; your audio console doesn’t have to be in one place anymore. It can be in 10 different places.”
Supported by
You might also like...
The Changing Face Of Live Sports: Part 1 - The Rise Of Nimble Production
Live sports broadcasting has always been the preserve of big leagues and big broadcasters with the infrastructure, the clout and the resources to match. But it is no longer the only game in town.
Standards: Audio - High Efficiency Audio Codecs (HE-AAC)
HE-AAC builds on the foundations of AAC to deliver near CD-quality audio at bitrates as low as 32 kbps, making it the codec of choice for mobile TV, digital radio and low-bandwidth streaming. This guide unpacks the key technologies behind its…
IP Security For Broadcasters 2026 – The Psychology Of Security
As engineers and technologists, it’s easy to become bogged down in the technical solutions that maintain high levels of computer security. But as the boundaries between traditional broadcast engineering and IT continue to dissolve, the first port of call i…
Standards: Audio - Advanced Audio Coding (AAC)
AAC succeeded MP3 by delivering better quality at lower bitrates. This guide examines how it works, compares the leading encoder implementations, and explains where it sits within the broader MPEG audio standards landscape.
Broadcast Standards - The Science Of AI: New Foundations
We begin this series with the foundational building blocks of AI. Basic principles, the technology stack and the types of AI based upon it, and how to apply them effectively in a broadcasting enterprise.