Tom Hartnett One year ago, Radio World asked several experts to discuss trends in the broadcast codec technology arena (Jan. 2, 2013). It was a popular topic, so we return to it now to learn more and find out what’s new in a set of four Q&As. Replies were via email and have been edited for length.
Here, we talk with Comrex Technical Director Tom Hartnett. Watch the website for the rest in the series.
What’s the most important recent change or improvement in broadcast codec technology?
Hartnett: The innovation in the telecom world to include “HD Voice” modes in non-broadcast products. Mobile phones, business desktop phones and softphone apps now all consider “telco quality” audio to be outdated, and are adding in high-fidelity codecs. Skype’s been doing this for years. Nobody is better positioned to benefit from that than radio. You just need studio hardware that can talk the right language.
What’s “state of the art” in algorithms?
Hartnett: AAC-ELD is still the winner for remotes and will continue to be for some time, as it provides the perfect balance between low network usage and low delay.
On its heels is Opus, which has only a slightly inferior sound but a big benefit for the telecom world in being free (as in speech and beer). Opus is being adopted quickly by Web developers. For non-delay intensive application (like streaming) HE-AAC is still king.
How good are low-bitrate codecs, and how much better can we expect performance to get?
Hartnett: The algorithms themselves are “good enough” now for the vast majority of applications. Innovation in this field has moved to things like error resiliency and variable bitrate operation on poor networks.
What relevance does the new AES67 standard have?
Hartnett: It’s early, and the new standard doesn’t include important elements like presence and discovery. So it’s not perfect. But from a product developer’s point of view it now opens up opportunity in more pure-software products. Most hardware codecs consist of analog and digital audio I/O tied to some type of internal computer. If we can now support the audio side via Ethernet (and remain AoIP vendor-agnostic), we can look at removing the custom hardware element completely and go pure software on a server. Especially for products that support lots of audio I/O channels, the idea of removing handfuls of large XLR connectors is very attractive. Of course, this only holds true for the studio side of any link, since field broadcasts are still tied to old-fashioned analog for now.
Assess the impact of the decline in POTS, ISDN and other older services on your users.
Hartnett: It can be frustrating for a broadcaster to have to change his paradigm, and expensive to buy new gear. But in the case of IP, it comes with one pretty cool feature that can be quite enabling: wireless over 3G/4G. If 4G coverage is good (and you’re not trying to cover an event with thousands of mobile phone users present), this becomes the ultimate broadcast tool — completely portable with zero setup, no installation cost and great audio. And 4G networks are getting better deployed every day.
What is your newest product or notable feature?
Hartnett: Our BRIC-Link Product, which is our entry-level simple hardware codec, is a huge seller for us. One of the many notable features of this box is the ability to push an Icecast-style stream to your webstream hosting company. We’re enhancing that aspect of the product by supporting an HTTP push mode that delivers FLAC (lossless) compression. This way the streaming company gets a perfect signal to start with before compressing to MP3 or AAC, eliminating any lossy transcoding. BRIC-Link can even act as a small-scale streaming server itself, delivering dozens of HE-AAC streams to listeners using web players directly.
Given advances in consumer audio and IP infrastructure, do you foresee a day when broadcasters don’t even need specialized “broadcast” codecs to move high-quality audio around?
Hartnett: We find that even when a consumer-grade product starts to shadow our offerings, many broadcasters will still need to rely on pro-grade gear with solid support. I’m thrilled that budget broadcasters are using Skype and similar stuff for remotes — first off, because I believe local presence is such an important aspect of the radio business, and cheap ways of achieving that makes radio better; but also because when they hit a certain threshold of success, they’ll call us for the “good stuff.”
One more important element in providing “codecs to consumers” is the ability for listeners to connect to the station in high fidelity. This is something we’re aggressively promoting with our STAC-VIP talkshow system. Radio will sound so much better when everyone calls in using codecs!
Comrex has spent a lot of time and resources on the video side of the industry. Will we see more of that?
Hartnett: Our LiveShot product does for video what we’ve done for audio, but that doesn’t make it a TV-only product. We’re finding the line is blurring, and it’s not an either/or thing. Radio stations are making the transition from audio providers to media providers via the Web, and often via cable TV.
There’s also a nice side effect of all the video development we’ve done in recent years. It turns out moving video over 4G networks is a lot more challenging than audio, due to the tenfold increase in network required. But we’ve licked that, and now we’re busy backporting all the R&D we’ve done to make our audio codecs more resilient on bad networks.
Anything else we should know?
Hartnett: At the risk of sounding a bit cagey, I’m really excited about the work being done by Web companies like Google in integrating real-time, high-quality media streaming into their products. I can see this stuff being available on every browser and on every computer or smartphone soon, and it gives radio an opportunity to forever rid itself of poor-quality telephone audio. We’ll be talking a lot more about that in 2014.
Comment on this or any story. Post below.