REMIX-CYCLE-CONSISTENT LEARNING ON ADVERSARIALLY LEARNED SEPARATOR FOR ACCURATE AND STABLE UNSUPERVISED SPEECH SEPARATION

Kohei Saijo, Tetsuji Ogawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

A new learning algorithm for speech separation networks is designed to explicitly reduce residual noise and artifacts in the separated signal in an unsupervised manner. Generative adversarial networks are known to be effective in constructing separation networks when the ground truth for the observed signal is inaccessible. Still, weak objectives aimed at distribution-to-distribution mapping make the learning unstable and limit their performance. This study introduces the remix-cycle-consistency loss as a more appropriate objective function and uses it to fine-tune adversarially learned source separation models. The remix-cycle-consistency loss is defined as the difference between the mixed speech observed at microphones and the pseudo-mixed speech obtained by alternating the process of separating the mixed sound and remixing its outputs with another combination. The minimization of this loss leads to an explicit reduction in the distortions in the output of the separation network. Experimental comparisons with multichannel speech separation demonstrated that the proposed method achieved high separation accuracy and learning stability comparable to supervised learning.

Original languageEnglish
Title of host publication2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4373-4377
Number of pages5
ISBN (Electronic)9781665405409
DOIs
Publication statusPublished - 2022
Event47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Virtual, Online, Singapore
Duration: 2022 May 232022 May 27

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2022-May
ISSN (Print)1520-6149

Conference

Conference47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
Country/TerritorySingapore
CityVirtual, Online
Period22/5/2322/5/27

Keywords

  • Deep neural networks
  • adversarial learning
  • remix-cycle-consistent learning
  • unsupervised speech separation

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'REMIX-CYCLE-CONSISTENT LEARNING ON ADVERSARIALLY LEARNED SEPARATOR FOR ACCURATE AND STABLE UNSUPERVISED SPEECH SEPARATION'. Together they form a unique fingerprint.

Cite this