Listen only to me! How well can target speech extraction handle false alarms?

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F22%3APU146143" target="_blank" >RIV/00216305:26230/22:PU146143 - isvavai.cz</a>

Result on the web

<a href="https://www.isca-speech.org/archive/pdfs/interspeech_2022/delcroix22_interspeech.pdf" target="_blank" >https://www.isca-speech.org/archive/pdfs/interspeech_2022/delcroix22_interspeech.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.21437/Interspeech.2022-11252" target="_blank" >10.21437/Interspeech.2022-11252</a>

Result language

angličtina

Original language name

Listen only to me! How well can target speech extraction handle false alarms?

Original language description

Target speech extraction (TSE) extracts the speech of a target speaker in a mixture given auxiliary clues characterizing the speaker, such as an enrollment utterance. TSE addresses thus the challenging problem of simultaneously performing separation and speaker identification. There has been much progress in extraction performance following the recent development of neural networks for speech enhancement and separation. Most studies have focused on processing mixtures where the target speaker is actively speaking. However, the target speaker is sometimes silent in practice, i.e., inactive speaker (IS). A typical TSE system will tend to output a signal in IS cases, causing false alarms. This is a severe problem for the practical deployment of TSE systems. This paper aims at understanding better how well TSE systems can handle IS cases. We consider two approaches to deal with IS, (1) training a system to directly output zero signals or (2) detecting IS with an extra speaker verification module. We perform an extensive experimental comparison of these schemes in terms of extraction performance and IS detection using the LibriMix dataset and reveal their pros and cons.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

R - Projekt Ramcoveho programu EK

Publication year

2022

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

ISBN

—

ISSN

1990-9772

e-ISSN

—

Number of pages

5

Pages from-to

216-220

Publisher name

International Speech Communication Association

Place of publication

Incheon

Event location

Incheon Korea

Event date

Sep 18, 2022

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—