Auxiliary loss function for target speech extraction and recognition with weak supervision based on speaker characteristics

Katerina Zmolikova*, Marc Delcroix, Desh Raj, Shinji Watanabe, Jan Černocký

*この研究の対応する著者

研究成果: Conference contribution

抄録

Automatic speech recognition systems deteriorate in presence of overlapped speech. A popular approach to alleviate this is target speech extraction. The extraction system is usually trained with a loss function measuring the discrepancy between the estimated and the reference target speech. This often leads to distortions to the target signal which is detrimental to the recognition accuracy. Additionally, it is necessary to have the strong supervision provided by parallel data consisting of speech mixtures and single-speaker signals. We propose an auxiliary loss function for retraining the target speech extraction. It is composed of two parts: first, a speaker identity loss, forcing the estimated speech to have correct speaker characteristics, and second, a mixture consistency loss, making the extracted sources sum back to the original mixture. The only supervision required for the proposed loss is speaker characteristics obtained from several segments spoken by the target speaker. Such weak supervision makes the loss suitable for adapting the system directly on real recordings. We show that the proposed loss yields signals more suitable for speech recognition and further, we can gain additional improvements by adaptation to target data. Overall, we can reduce the word error rate on LibriCSS dataset from 27.4% to 24.0%.

本文言語English
ホスト出版物のタイトル22nd Annual Conference of the International Speech Communication Association, INTERSPEECH 2021
出版社International Speech Communication Association
ページ4156-4160
ページ数5
ISBN(電子版)9781713836902
DOI
出版ステータスPublished - 2021
外部発表はい
イベント22nd Annual Conference of the International Speech Communication Association, INTERSPEECH 2021 - Brno, Czech Republic
継続期間: 2021 8月 302021 9月 3

出版物シリーズ

名前Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
6
ISSN(印刷版)2308-457X
ISSN(電子版)1990-9772

Conference

Conference22nd Annual Conference of the International Speech Communication Association, INTERSPEECH 2021
国/地域Czech Republic
CityBrno
Period21/8/3021/9/3

ASJC Scopus subject areas

  • 言語および言語学
  • 人間とコンピュータの相互作用
  • 信号処理
  • ソフトウェア
  • モデリングとシミュレーション

フィンガープリント

「Auxiliary loss function for target speech extraction and recognition with weak supervision based on speaker characteristics」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル