As a result of increasing computational performance and dramatically lower cost, embedded systems have become suitable for digital signal processing of audio signals. This paper describes a novel multichannel, low-latency, Linux-based audio system that is appropriate for real-time processing. Two different driver architectures are described: a common Linux architecture using ALSA and a unique real-time driver architecture (Bela Platform). The development of the ALSA driver architecture includes device drivers that use the ASoC layer, sound server settings, device tree overlays and capes, register maps, and real-time patches to the kernel. The adaption of the multichannel sound card drivers for Bela focuses on a hard real-time program for data transmission, multichannel buffer alignments, and audio codec control. The overall system has been evaluated with respect to sound quality and latency to gauge its usefulness as a powerful new platform for audio development projects, such as embedded digital effect processors for musicians or augmented and participatory performances.
(See document for exact affiliation information.)
https://aes2.org/publications/elibrary-page/?id=19394
(391KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
