Download the presentation made at the Embedded Linux Conference 2011 in San Francisco, CA (USA) by Remi Lorriaux, one of our embedded software engineer.
Overview of the Presentation
This presentation will first describe the challenges that programmers face when dealing with real-time sound processing on embedded devices. It will then go through the steps that can be taken to improve the audio latency while preserving the sound quality. Starting from a general-purpose embedded Linux platform, it will show how the kernel and applications can be modified and written to meet these constraints. A solution involving the use of a DSP in conjunction with an ARM CPU will also be discussed. This presentation is aimed at developers working on systems where audio performance is critical. This includes for instance voice processing, communications, music production, etc... While a lot of information is already available in the community for desktop platforms, the topic is rarely discussed in the embedded world, where low power consumption and resource usage are essential.
The Remi's presentation is available from this webpage
You can also find other presentations on the official page collecting the presentations that were made at the conference: http://elinux.org/ELC_2011_Presentations