🚀 go-pugleaf

> Subject: 300 Baud Speech Vocoding...
>
> I'm a little surprised no one has talked about the low data-rate
> speech vocoding algorithms that have been developed.

I am also interested in speech encoding algorithms.

Specifically, I'm interested in low- to medium-rate algorithms (300 bps to
16 kbps) that have been implemented for real-time operation on
off-the-shelf hardware (e.g., on the NeXT computer, or on DSP boards for
the SPARCstation or the PC).  I'd greatly appreciate receiving any
information on such implementations that could be obtained with little
cost.  (I realize that understandable-quality, real-time, < 1 kbps
encoding, on off-the-shelf hardware is probably asking too much.)

For context: In a current research project, we are building a speech
interface to variable data rate packet-switched networks (e.g., packet
radio networks) that can adapt its encoding rate, possibly by
switching between different encoding algorithms, according to the
current throughput provided by the underlying network.  Of course, any
"software contributions" would be acknowledged.

Thanks in advance for any information,

Dave Beyer
SRI International
Menlo Park, CA  94025
beyer@sri.com
(415) 859-6335