San Francisco - Nov 5, 2010 - CSR demonstrated a DSP-based evaluation platform for its low-latency apt-X Live codec at the 2010 AES Convention. Helping overcome the spectrum efficiency and audio quality challenges involved in digital wireless microphone and in-ear monitoring system design, CSR's evaluation platform demonstrated audio bandwidth, dynamic range and audio delay performance.

An important attraction of digital wireless microphone systems is their potential to offer more efficient use of the radio frequency spectrum for a given audio quality. Key to microphone designs achieving spectrum efficiency is the use of low-latency audio compression technology as an essential sub-system in the signal processing chain.

CSR's apt-X Live codec is specifically designed for the requirements of digital wireless microphones. The patented architecture provides audio data rate reduction and an enhanced degree of error tolerance, all with extremely low latency.

This apt-X Live evaluation platform comprised separate digital radio transmitter and receiver modules and was designed to demonstrate how the low-latency audio codec subsystem within a typical digital wireless microphone system can be successfully implemented.

In operation, the programmable DSP-based digital radio transmitter module takes a mono channel audio input (16-bit resolution, sampled at 48kHz) and compresses this in real-time using the apt-X Live algorithm running at 2.5 bits per PCM sample to produce a compressed bitstream at 120kbits/sec. This bitstream is digitally modulated using FSK onto an 868MHz RF carrier, and transmitted over the air to a similar DSP-based receiver module that demodulates the bitstream and decompresses the data to reproduce the original PCM audio signal.

The apt-X Live audio codec can deliver a single channel of 24bit, 48kHz audio in 120kb/s with a codec latency of less than 1.9ms .The framework of apt-X Live audio coding will provide a higher resilience to random bit errors. Additional techniques employed within apt-X Live also will reduce the significance of these errors and the magnitude of the audible effect on certain signals.