Exploiting the Intermittency of Speech for Joint Separation and Diarization - Archive ouverte HAL Access content directly
Conference Papers Year : 2017

Exploiting the Intermittency of Speech for Joint Separation and Diarization

(1) , (2, 1) , (1) , (1) , (3)
1
2
3

Abstract

Natural conversations are spontaneous exchanges involving two or more people speaking in an intermittent manner. Therefore one expects such conversation to have intervals where some of the speakers are silent. Yet, most (multichannel) audio source separation (MASS) methods consider the sound sources to be continuously emitting on the total duration of the processed mixture. In this paper we propose a probabilistic model for MASS where the sources may have pauses. The activity of the sources is modeled as a hidden state, the diarization state, enabling us to activate/de-activate the sound sources at time frame resolution. We plug the diarization model within the spatial covariance matrix model proposed for MASS, and obtain an improvement in performance over the state of the art when separating mixtures with intermittent speakers.
Fichier principal
Vignette du fichier
main.pdf (2.48 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01568813 , version 1 (25-07-2017)

Identifiers

Cite

Dionyssos Kounades-Bastian, Laurent Girin, Xavier Alameda-Pineda, Radu Horaud, Sharon Gannot. Exploiting the Intermittency of Speech for Joint Separation and Diarization. WASPAA 2017 - IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, Oct 2017, New Paltz, NY, United States. pp.41-45, ⟨10.1109/WASPAA.2017.8169991⟩. ⟨hal-01568813⟩
245 View
250 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More