[orca-list] audio mixing and speech responsiveness

[Jason White <jason jasonjgw net>]

On Mon, Dec 01, 2008 at 11:55:13PM +0000, Michael Whapples wrote:
> Regarding the responsiveness of orca, I believe this is down to ubuntu  
> using pulseaudio but OpenSolaris doesn't. I believe when I have got a  
> Linux system (slackware) using ALSA only then responsiveness seems to be  
> reasonably the same.

On my Debian systems, which use Alsa only (no Pulse Audio daemon), I notice
that when multiple Alsa applications are active simultaneously, the sound is
mixed properly. However, if I run an application that uses OSS while an Alsa
application is writing to the audio output, the OSS application receives a
"device or resource busy" error.

My solution to this is to avoid OSS applications. They're mostly legacy
programs that haven't been updated for years, and for which newer alternatives
exist that take advantage of Alsa. It helps that I don't run proprietary
speech synthesis engines, some of which use OSS.

There is supposed to be a solution to the OSS/Alsa audio mixing problem that
can be implemented in your ~/.asoundrc file, but I haven't been able to get it
to work. Neither have I devoted much effort to it, however.

http://alsa.opensrc.org/DmixPlugin