Speaker diarization using unsupervised discriminant analysis of inter-channel delay features

Abstract : When multiple microphones are available estimates of inter-channel delay, which characterise a speaker's location, can be used as features for speaker diarization. Background noise and reverberation can, however, lead to noisy features and poor performance. To ameliorate these problems, this paper presents a new approach to the discriminant analysis of delay features for speaker diarization. This novel and nonetheless unsupervised approach aims to increase speaker separability in delay-space. We assess the approach on subsets of four standard NIST RT datasets and demonstrate a relative improvement in diarization error rate of 25% on a separate evaluation set using delay features alone.
Type de document :
Communication dans un congrès
IACSSP'09, Apr 2009, Taipei, Taiwan. 2009, 〈10.1109/ICASSP.2009.4960520〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01318388
Contributeur : Bibliothèque Universitaire Déposants Hal-Avignon <>
Soumis le : jeudi 19 mai 2016 - 15:23:58
Dernière modification le : dimanche 10 février 2019 - 10:32:05

Identifiants

Collections

Citation

Nicholas Evans, Corinne Fredouille, Jean-François Bonastre. Speaker diarization using unsupervised discriminant analysis of inter-channel delay features. IACSSP'09, Apr 2009, Taipei, Taiwan. 2009, 〈10.1109/ICASSP.2009.4960520〉. 〈hal-01318388〉

Partager

Métriques

Consultations de la notice

48