home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!well!moon!rds!doug
- From: doug@rds.com (Doug Kaye)
- Newsgroups: comp.speech
- Subject: Compression Divergence
- Message-ID: <3wm4TB1w164w@rds.com>
- Date: Wed, 11 Nov 92 15:54:13 PST
- Distribution: world
- Organization: Rational Data Systems, San Rafael, CA
- Lines: 20
-
- Normally (ie, in a real-time enviornment) audio is compressed then
- decompressed and that's it. However in certain non-real-time
- applications audio may be compressed and decompressed repeatably. In
- such applications it is important that the compression/decompression
- algorithms be "convergent". By this, I mean that within a small number
- of compress/decompress cycles the distortion ceases to increase.
-
- We are currently using Intel's DVI ADPCM algorithm because (among its
- other advantages) it converges after the very first such cycle.
- Successive compress/decompress cycles result in identical data.
-
- Has anyone had experience with regard to this convergence pheonomenon
- with CELP? VSELP? MPEG? Other audio compression algorithms?
-
- ...doug
-
- ==================================================
- Doug Kaye Rational Data Systems San Rafael, CA
- doug@rds.com {moon|uunet}!rds!doug MHS:doug@rds
- ==================================================
-