🚀 go-pugleaf

I'm looking for some good sound compression and silence detection
algorithms for a sun. So far, I've gotten the adpcm stuff and gotten
that to work on the sun. It seems okay, and I've been able to cut
the volume in half using that by translating from Ulaw to signed
16 and then doing the 16 -> 4bit quantization. However, I'm afraid
that all the conversions that are involved are going to slow up
my throughput as I plan on doing the conversion on the fly for 
packet voice.  I guess what I really need is to somehow detect when
the incoming volume on the audio port is below a certain threshhold
and then not send packets over the net until it breaks above that
threshold again.  It will probably mean smaller packets, but so be-it.
   I've also gotten the CELP stuff.. But, the documentation only tells
you how to compile the stuff.  It does not say what it should be used for.
I haven't really heard anything about CELP so I don't know what kind
of use I can get out of it or if it would be good for voice compression
on the fly.  It says something about an input file also, but it doesn't
really explain what the input file should be.
   Any help would be appreciated! Thanks!

-- 
 ___			|	GE software developer
( / ) _       _,	|	doug@happy.vf.ge.com
_/_/ <_>_/_/_<_>	|	uunet!crdgw1!ge-dab!happy.vf.ge.com!doug
             <_>	|	2B | !2B > The_Question
________________________|______________________________________________________