Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

MULTICHANNEL SPEECH ENHANCEMENT FOR SPEAKER VERIFICATION IN NOISY AND REVERBERANT ENVIRONMENTS

Sandipana Dowerah 1 Romain Serizel 1 Denis Jouvet 1 Mohammad Mohammadamini 2 Driss Matrouf 2
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Speech signals can be corrupted by environmental noise as well as room reverberation which severely affects the speaker verification performance. In this paper, we propose to combine a multichannel pre-processing pipeline including filter-and-sum network (FaSnet), Rank-1 multichannel Wiener filter, and weighted prediction error as a front-end to speaker verification. Experimental evaluation shows that the pre-processing can improve the speaker verification performance as long as the enrollment files are processed similarly to the test data and that test and enrollment occur within similar SNR ranges. Our proposed pipeline is trained on synthetic data but generalizes to unseen, real recorded clips included in the VOiCES eval dataset and improves the speaker verification performance on all the noise conditions.
Document type :
Preprints, Working Papers, ...
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03487420
Contributor : Sandipana Dowerah Connect in order to contact the contributor
Submitted on : Friday, December 17, 2021 - 3:44:51 PM
Last modification on : Tuesday, January 4, 2022 - 6:20:01 AM

File

ICASSP_22.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03487420, version 1

Citation

Sandipana Dowerah, Romain Serizel, Denis Jouvet, Mohammad Mohammadamini, Driss Matrouf. MULTICHANNEL SPEECH ENHANCEMENT FOR SPEAKER VERIFICATION IN NOISY AND REVERBERANT ENVIRONMENTS. 2021. ⟨hal-03487420⟩

Share

Metrics

Les métriques sont temporairement indisponibles