Sökning: id:"swepub:oai:DiVA.org:kth-301123" >
The CAVE-WP4 generi...
The CAVE-WP4 generic speaker verification system
-
- Jaboulet, C. (författare)
- IDIAP, Rue du Simplón 4, Case Postale 592, Martigny, CH-1920, Switzerland ; Ubilab, Union Bank of Switzerland, Bahnhofstrasse 45, Zürich, CH-8021, Switzerland
-
- Koolwaaij, J. (författare)
- KUN, Dept of Language and Speech, Erasmusplein 1, Nijmegen, NL-6525 HT, Netherlands
-
- Lindberg, Johan (författare)
- KTH,Tal, musik och hörsel, TMH
-
visa fler...
-
- Pierrot, J. -B (författare)
- ENST, Dépt Signal, CNRS, URA 820, 46 Rue Barrault, Paris Cedex 13, 75634, France
-
- Bimbot, F. (författare)
- ENST, Dépt Signal, CNRS, URA 820, 46 Rue Barrault, Paris Cedex 13, 75634, France
-
visa färre...
-
(creator_code:org_t)
- International Speech Communication Association, 1998
- 1998
- Engelska.
-
Ingår i: RLA2C 1998 - Speaker Recognition and its Commercial and Forensic Applications. - : International Speech Communication Association. ; , s. 202-205
- Relaterad länk:
-
https://urn.kb.se/re...
Abstract
Ämnesord
Stäng
- This papers describes the generic speaker verification system that was developed within Work Package 4, which was responsible for Research and Technology improvements for Speaker Verification in the CAVE project. With this perspective, different SV algorithms were implemented in a common HMM framework and compared on several databases. This paper is concerned with the design of this generic system and not with the experiments we performed using the system. It shows the importance of having a common and generic package for cooperative research. We intend to make the Generic System available to the research community.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Språkteknologi (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Language Technology (hsv//eng)
Nyckelord
- Caves
- Digital forensics
- Generic system
- Research communities
- Speaker verification
- Speaker verification system
- Technology improvement
- Work packages
- Speech recognition
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)