Tests with emacspeak and 4front soft mixer

["T. V. Raman" <ramantv@xxxxxxxxxxx>]

without looking at the mixer code to see what it is doing
it's hard to say --but here is an educated guess as to why
you're seeing processor load even when it is not playing
multiple streams:

In order to mix multiple streams I suspect they are
normalizing all incoming streams to the same sample rate.
ESD (Enlightenment Sound Daemon) for instance takes
everything to 44k --guess what --that will cost you
computes.

In fact if you start up esd and play an 8k ulaw file you
will notice degradation on slow systems.

The Viavoice TTS engine produces 16K speech by default 
--so if they are upsampling that to 44k you will definitely
take a hit.
The above sounds probable given you say you heard the same
word repeated.

The state of audio on Linux is unfortunately pretty sad --at
this point in time even the darkside does better.  It's
particularly sad because these problems were solved a long
time ago for the most part in projects like AudioFile (AF)
--a client/server audio system based on the X server
architecture and codebase.

At some point sound on Linux will catch up --but I suspect
for now it's "good enough" for what the major projects want
--and moreover there are more  pressing problems 
in other corners of Linux.

Wait for gaming to become popular on Linux before you see
these fixed --OReillyNet recently ran a very nice article on
3Daudio and gaming on Linux 
--might be hope left still.

-- 
Best Regards,
--raman

      
Email:  raman@xxxxxxxxxxx
WWW: http://www.cs.cornell.edu/home/raman/             
PGP:    http://www.cs.cornell.edu/home/raman/raman.asc 

-----------------------------------------------------------------------------
To unsubscribe from the emacspeak list or change your address on the
emacspeak list send mail to "emacspeak-request@xxxxxxxxxxx" with a
subject of "unsubscribe" or "help"