//php echo do_shortcode(‘[responsivevoice_button voice=”US English Male” buttontext=”Listen to Post”]’) ?>
Ultra-low latency, digital wireless audio is the holy grail for a range of sound and music applications, including those in live performance and production, gaming and Extended reality. While the advantages of freeing artists, gamers, users and their devices from cables are clear, the performance needs of these applications have not been met by existing wireless technologies.
Audio latency—the time delay in audio introduced by its transit through the wireless system—is the main challenge. For passionate users in these markets, this parameter is critical to their experience—and therefore to purchasing decisions.
Of course, other considerations, including range and reliability, battery lifetime and device cost, factor into the decision making—but only if the latency is acceptable.
Ultra-low latency means a total system delay (from analog to analog) that is significantly below 10 milliseconds. For live feedback and monitoring applications, the total delay from instrument or microphone back to the earpiece (or other monitor) needs to be below this duration or else the performance begins to suffer due to the disconnect between the artist’s actions and what they hear. (As an aside, it is interesting to note that 10 milliseconds is about the time it takes sound to propagate at room temperature over about 3½ meters, or 11 feet. That is about the same width as a typical garage- or pub-band setup, which is not a coincidence.)
Bearing in mind that in live performance scenarios the audio often also must traverse a mixing desk and possibly a digital effects system, this leaves the wireless system needing to achieve a latency below 5 milliseconds to meet the application needs.
This level of performance can be achieved in analog wireless systems that often use sub-GHz UHF frequencies to directly and continuously modulate the RF carrier using the audio signal. But these systems have limitations in areas where digital can outperform. Digital systems can:
- Make more efficient use of spectrum, enabling many simultaneous collocated non-interfering links, possibilities for multi-channel audio transport, and use of the system in global bands where regulatory constraints require high-efficiency modulations;
- Through quantization, coding and protocol, better sustain a high signal-to-noise ratio for the audio across the range of RF channel conditions from optimal to marginal;
- Support the use of cryptographic techniques to provide privacy and/or authenticity of the content;
- Enable flexible network topologies, for example, point-to-point, point-to-multipoint and multipoint-to-point, and
- Provide multiplexed data side-channels to support content-related metadata or control/synchronization traffic.
But general-purpose, digital wireless systems aren’t typically well suited to audio transport. The advantages above are achieved through mechanisms like packetization, audio encoding/compression and link-layer retransmission schemes that introduce undesirable delay into the link.
This is demonstrated, for example, in the Bluetooth profiles that provide for audio transport. While the Bluetooth Classic physical and link layers support high data rates (up to 3 Mbps) and relatively fast packet transmission rates (up to one per 1.25 ms), the best system latency achievable with the full protocol stack is only around 30 ms for voice-quality audio via HFP, and 100-150 ms for stereo audio via A2DP.
Advanced codecs, such as Qualcomm’s aptX Low Latency, significantly improve the latter to around 40 ms. However, this is still an order of magnitude higher than would be tolerable for many ultra-low-latency applications.
Bluetooth LE Audio does better, but not by far. By way of a new low-latency isochronous link layer protocol, and use of the Low Complexity Communication Codec (LC3) developed by Fraunhofer IIS and Ericsson, LE Audio systems can deliver stereo audio latencies only down to around 20-30 ms.
So, how can we have the best of both worlds? How can we marry the precision, flexibility and scalability of digital and the market-leading low latency of analog systems?
This is a question the team at Virscient has focused on over the last few years. We believe the answer has several aspects, namely:
- Building on standards is critical. Technologies like Bluetooth LE provide strong interoperability with modern smartphones, and include mechanisms for secure pairing, key distribution and link coordination. The Bluetooth LE physical layer can support data rates for (small numbers of) multi-channel high-quality audio if advanced codecs are used. Where highly multi-channel or lossless audio is required, or long-range transmission needed, alternative RF transports like UWB can be used as a sideband.
- Careful codec selection is needed to optimise latency vs bandwidth. Typical low-complexity or high-compression codecs (especially those optimized for low-power devices) often have relatively long codec delay. Use of high-performance codecs like Skylark by Audio Codecs, can significantly improve end-to-end latency whilst maintaining broadcast-quality audio.
- Audio-centric protocols and close integration between layers of the implementation are essential to reducing latency. In particular, optimized link-layer protocols ensure audio stream robustness with RF in licensed or unlicensed spectrum, with minimum latency compromise.
We intend to enable a range of long-awaited applications for live audio and gaming, on hardware platforms with a low BoM cost and high performance. As a team of engineers, developers, musicians, gamers and audio enthusiasts, we can’t wait to see what people will build with it.
