AVX Architect - What is Acoustic Echo Cancellation (AEC)?
Table of Contents
Brands:
- AMX
- BSS Audio
Models:
- AVX Architect
Overview:
This article continues the exploration of the VIRTUAL view in AVX Architect and explains Acoustic Echo Cancellation (AEC).
To see a video of this article, click HERE.
Description:
Before reviewing applications of AEC, it is necessary to define some basic terms and describe how AEC functions.
Acoustic echo occurs in a conferencing system when far-side speech played through local loudspeakers is picked up by microphones in the near-side room and transmittedBack to the far side.
This transmitted signal is a delayed version of the original, which causes the echo for the far side participant. The received far-side signal does not transfer directly from the loudspeaker to the microphone but is affectedBy room artifacts. These artifacts may include differing signal paths that cause reverberation, frequency filtering, and attenuation. These effects are the result of the transfer function of the room. The transfer function is dynamic because objects in the room may move or the microphone position may change. To subtract the unwanted signal correctly, the AEC processor must simulate the dynamic room transfer function. The processor then applies that transfer function to the received signal and subtracts the modified original signal.
The OMNI 256p processor provides 80 channels of AEC processing, while the OMNI 512p processor provides 160 channels.
Each channel of processing includes a 48k algorithm, individual or shared AEC references, echo reduction of 40 dB or more, echo suppression of up to 80 dB, noise reduction up to 25 dB, adaptive non-linear processing, and convergence rates of 100 dB per second, resulting in only 16 ms of latency.
Review the following AEC terms and definitions before applying them in a configuration:
- Acoustic Echo Cancellation (AEC): Removes the far-side signal picked up by in-room microphones from the signal transmittedBack to the far side.
- Convergence Rate: Measures how quickly the AEC algorithm can identify and remove echo from the signal path.
- Echo Return Loss (ERL): Indicates the coupling between the AEC reference signal and the AEC input signal.
- Echo Return Loss Enhancement (ERLE): Shows the loss through the linear AEC algorithm, excluding non-linear processing.
- Noise Cancelling (NC): Also referred to as noise reduction, removes unwantedBackground signals.
- Residual Echoes: Echoes that may remain due to real-world acoustic conditions and limitations of the linear processing filter.
- Non-Linear Processing Stage: Further suppresses residual echoes in challenging acoustic environments and enhances performance during double-talk situations when both near-end and far-end speakers are talking simultaneously.
Table of Contents