5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Speaker Verification in the Telephone Network: Research Activities in the Cave Project

Frédéric Bimbot (1), Hans-Peter Hutter (5), Cedric Jaboulet (2), Johan W. Koolwaaij (4), Johan Lindberg (3), Jean-Benoit Pierrot (1)

(1) ENST / CNRS, Paris, France; (2) IDIAP, Martigny, Switzerland; (3) KTH, Stockholm, Sweden; (4) KUN, Nijmegen, The Netherlands; (5) Ubilab-UBS, Switzerland

This paper summarizes the main results from the Speaker Verification (SV) research pursued so far in the CAVE project. Different state-of the art SV algo- rithms were implemented in a common HMM frame- work and compared on two databases : YOHO (of fice environment speech) and SESP (telephone speech). This paper is concerned with the different design issues for LR-HMM-based SV algorithms which emerged from our investigations and which led to our current SV sys- tem, which delivers Equal Error Rates below 0.5 % on a very realistic telephone speech database.

Full Paper

Bibliographic reference.  Bimbot, Frédéric / Hutter, Hans-Peter / Jaboulet, Cedric / Koolwaaij, Johan W. / Lindberg, Johan / Pierrot, Jean-Benoit (1997): "Speaker verification in the telephone network: research activities in the cave project", In EUROSPEECH-1997, 971-974.