home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!gatech!hubcap!chinook!bdbryan
- From: bdbryan@eng.clemson.edu (Ben Bryant)
- Subject: Temporal networks
- Message-ID: <1992Aug13.125946.3197@hubcap.clemson.edu>
- Sender: news@hubcap.clemson.edu (news)
- Reply-To: bdbryan@eng.clemson.edu
- Organization: College of Engineering, Clemson Univ.
- Date: Thu, 13 Aug 1992 12:59:46 GMT
- Lines: 17
-
- I have recently been doing some work with speech recognition, and have used a
- few network configurations in my search for "the perfect network architecture".
- I wanted to find out from people in this group what some of the best networks
- for sequence recognition are? -A few examples are: TDNN's (Dr. Waibel at CMU),
- Spatio-Temporal Nets (-The author's names slip my mind currently), and a
- version of Kohonen's SOFMs that use multiple frames per vector (this gives
- some temporal invariance). -Are there others that you know about? -If you
- would be so kind as to mail the name of the architecture and the inventor
- to me - along with a reference or two on that architecture, I will be glad
- to post a summary of responses in a few weeks. -Thanks much.
-
- -Benjamin Bryant
- <bdbryan@eng.clemson.edu>
-
-
-
-
-