Speaker activity driven neural speech extraction

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F21%3APU140966" target="_blank" >RIV/00216305:26230/21:PU140966 - isvavai.cz</a>

Result on the web

<a href="https://www.fit.vut.cz/research/publication/12479/" target="_blank" >https://www.fit.vut.cz/research/publication/12479/</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICASSP39728.2021.9414998" target="_blank" >10.1109/ICASSP39728.2021.9414998</a>

Result language

angličtina

Original language name

Speaker activity driven neural speech extraction

Original language description

Target speech extraction, which extracts the speech of a target speaker in a mixture given auxiliary speaker clues, has recently received increased interest. Various clues have been investigated such as pre-recorded enrollment utterances, direction information, or video of the target speaker. In this paper, we explore the use of speaker activity information as an auxiliary clue for single-channel neural network-based speech extraction. We propose a speaker activity driven speech extraction neural network (ADEnet) and show that it can achieve performance levels competitive with enrollmentbased approaches, without the need for pre-recordings. We further demonstrate the potential of the proposed approach for processing meeting-like recordings, where speaker activity obtained from a diarization system is used as a speaker clue for ADEnet. We show that this simple yet practical approach can successfully extract speakers after diarization, which leads to improved ASR performance when using a single microphone, especially in high overlapping conditions, with relative word error rate reduction of up to 25 %.

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

<a href="/en/project/LQ1602" target="_blank" >LQ1602: IT4Innovations excellence in science</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2021

Confidentiality

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

Article name in the collection

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ISBN

978-1-7281-7605-5

ISSN

—

e-ISSN

—

Number of pages

5

Pages from-to

6099-6103

Publisher name

IEEE Signal Processing Society

Place of publication

Toronto

Event location

Toronto, Canada

Event date

Jun 6, 2021

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

000704288406074