Newsgroups: comp.speech
Path: pavo.csi.cam.ac.uk!doc.ic.ac.uk!warwick!uknet!pipex!uunet!gatech!howland.reston.ans.net!newsserver.jvnc.net!newsserver.technet.sg!solomon.technet.sg!alexlee
From: alexlee@solomon.technet.sg (Dr Alex Lee)
Subject: Re: speech compression
Message-ID: <CA7tJx.5oL@newsserver.technet.sg>
Sender: news@newsserver.technet.sg
Nntp-Posting-Host: solomon.technet.sg
X-Newsreader: TIN [version 1.2 PL0]
References: <20uorg$6ek@vtserf.cc.vt.edu> <2176jkINNfj6@master.cs.rose-hulman.edu> <C9vJF1.M87@newsserver.technet.sg> <jimn8C9vsvK.2E5@netcom.com>
Date: Thu, 15 Jul 1993 16:52:43 GMT
Lines: 40

Jim Nitchals (jimn8@netcom.com) wrote:
: alexlee@solomon.technet.sg (Dr Alex Lee) writes:

: >I have the CELP code compiled and in working condition on a DecStation
: >running under Ultrix.  I also have a PC version compiled with Borland
: >C and a faster version compiled with gcc (djgpp).

: Where did you get it?  I checked furmint.nectar.cs.cmu.edu, and the
: promised celp.audio.compression directory wasn't there; just a file
: with notes.


I got the source from furmint.nectar.  celp.audio.compression WAS a
directory with the source codes and samples.  I just checked the
site, and as you metioned, the directory has been replaced with a file
with notes.

Following my post, I received a lot of help from Joe Cambell and Ed Hall
to figure out why the quality of the speech files I compressed on my
PC system was so poor.  We suspect two possible reasons:  one, poor
handling of the analogue portion of my recording, and two, something
wrong with compatibility between the compression code (ie ANALYSIS)
and the compilers used.

I'm quite confident now the decompression part is OK (ie SYNTHESIS),
as a sample sent to me by Ed decompressed beautifully on my PC.

I checked the DecStation/Ultrix compiled code, and found that the
speech quality on the compression side was similar to the PC side.
As I compiled with gcc on the Ultrix system (as in the PC system),
the problem might lie with imcompatibilities between Sun cc and
gcc.  Another possibility is some errors introduced inadvertantly
in the compression side of the souce code during my porting effort.

I suggest you email Joe Campbell and ask him to make the source
available again on furmint or another site.  Meanwhile, I'm
still investigating my code.

Alex

