A Lightweight Speech Detection System for Perceptive Environments

Dominique Vaufreydaz 1 Rémi Emonet 1 Patrick Reignier 1
1 PRIMA - Perception, recognition and integration for observation of activity
Inria Grenoble - Rhône-Alpes, UJF - Université Joseph Fourier - Grenoble 1, INPG - Institut National Polytechnique de Grenoble , CNRS - Centre National de la Recherche Scientifique : UMR5217
Abstract : In this paper, we address the problem of speech activity detection in multimodal perceptive environments. Such space may contain many different microphones (lapel, distant or table top). Thus, we need a generic speech activity detector in order to cope with different speech conditions (from closetalking to noisy distant speech). Moreover, as the number of microphones in the room can be high, we also need a very light system. The speech activity detector presented in this article works efficiently on dozens of microphones in parallel. We will see that even if its absolute score of the evaluation is not perfect (30% and 40% of error rate respectively on the two tasks), its accuracy is good enough in the context we are using it.
Document type :
Conference papers
Complete list of metadatas

Cited literature [11 references]  Display  Hide  Download

https://hal.inria.fr/inria-00326529
Contributor : Dominique Vaufreydaz <>
Submitted on : Friday, October 3, 2008 - 12:41:14 PM
Last modification on : Tuesday, November 27, 2018 - 8:46:04 PM
Long-term archiving on : Friday, June 4, 2010 - 12:10:18 PM

File

Vaufreydaz06.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00326529, version 1

Collections

Citation

Dominique Vaufreydaz, Rémi Emonet, Patrick Reignier. A Lightweight Speech Detection System for Perceptive Environments. 3rd Joint Workshop on Multimodal Interaction and Related Machine Learning Algorithms, May 2006, Washington, United States. ⟨inria-00326529⟩

Share

Metrics

Record views

303

Files downloads

179