End-to-end speaker segmentation for overlap-aware resegmentation - Le Mans Université Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

End-to-end speaker segmentation for overlap-aware resegmentation

Résumé

Speaker segmentation consists in partitioning a conversation between one or more speakers into speaker turns. Usually addressed as the late combination of three sub-tasks (voice activity detection, speaker change detection, and overlapped speech detection), we propose to train an end-to-end segmentation model that does it directly. Inspired by the original end-to-end neural speaker diarization approach (EEND), the task is modeled as a multi-label classification problem using permutation-invariant training. The main difference is that our model operates on short audio chunks (5 seconds) but at a much higher temporal resolution (every 16ms). Experiments on multiple speaker diarization datasets conclude that our model can be used with great success on both voice activity detection and overlapped speech detection. Our proposed model can also be used as a post-processing step, to detect and correctly assign overlapped speech regions. Relative diarization error rate improvement over the best considered baseline (VBx) reaches 17% on AMI, 13% on DIHARD 3, and 13% on VoxConverse.
Fichier principal
Vignette du fichier
2104.04045.pdf (438.9 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03257524 , version 1 (11-06-2021)

Identifiants

Citer

Hervé Bredin, Antoine Laurent. End-to-end speaker segmentation for overlap-aware resegmentation. Interspeech 2021, Aug 2021, Brno, Czech Republic. ⟨hal-03257524⟩
222 Consultations
176 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More