Filterbank design for end-to-end speech separation

Manuel Pariente 1 Samuele Cornell 2 Antoine Deleforge 1 Emmanuel Vincent 1
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Single-channel speech separation has recently made great progress thanks to learned filterbanks as used in ConvTasNet. In parallel, parameterized filterbanks have been proposed for speaker recognition where only center frequencies and bandwidths are learned. In this work, we extend real-valued learned and parameterized filterbanks into complex-valued analytic filterbanks and define a set of corresponding representations and masking strategies. We evaluate these fil-terbanks on a newly released noisy speech separation dataset (WHAM). The results show that the proposed analytic learned filterbank consistently outperforms the real-valued filterbank of ConvTasNet. Also, we validate the use of parameterized filterbanks and show that complex-valued representations and masks are beneficial in all conditions. Finally, we show that the STFT achieves its best performance for 2 ms windows.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02355623
Contributor : Manuel Pariente <>
Submitted on : Friday, November 8, 2019 - 3:15:30 PM
Last modification on : Saturday, November 9, 2019 - 2:14:41 AM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02355623, version 1

Citation

Manuel Pariente, Samuele Cornell, Antoine Deleforge, Emmanuel Vincent. Filterbank design for end-to-end speech separation. 2019. ⟨hal-02355623⟩

Share

Metrics

Record views

22

Files downloads

13