Publication Details

Neural Target Speech Extraction: An overview

ŽMOLÍKOVÁ, K.; DELCROIX, M.; OCHIAI, T.; ČERNOCKÝ, J.; KINOSHITA, K.; YU, D. Neural Target Speech Extraction: An overview. IEEE SIGNAL PROCESSING MAGAZINE, 2023, vol. 40, no. 3, p. 8-29. ISSN: 1558-0792.
Czech title
Neurální extrakce řeči cílového mluvčího: Přehled
Type
journal article
Language
English
Authors
Žmolíková Kateřina, Ing., Ph.D. (FIT)
Delcroix Marc
OCHIAI, T.
Černocký Jan, prof. Dr. Ing. (DCGM)
Kinoshita Keisuke
Yu Dong
URL
Keywords

neural, speech, extraction

Abstract

Humans can listen to a target speaker even in challenging acoustic conditions
that have noise, reverberation, and interfering speakers. This phenomenon is
known as the cocktail party effect . For decades, researchers have focused on
approaching the listening ability of humans. One critical issue is handling
interfering speakers because the target and nontarget speech signals share
similar characteristics, complicating their discrimination. Target speech/speaker
extraction (TSE) isolates the speech signal of a target speaker from a mixture of
several speakers, with or without noises and reverberations, using clues that
identify the speaker in the mixture. Such clues might be a spatial clue
indicating the direction of the target speaker, a video of the speaker's lips,
and a prerecorded enrollment utterance from which the speaker's voice
characteristics can be derived. TSE is an emerging field of research that has
received increased attention in recent years because it offers a practical
approach to the cocktail party problem and involves such aspects of signal
processing as audio, visual, and array processing as well as deep learning. This
article focuses on recent neural-based approaches and presents an in-depth
overview of TSE. We guide readers through the different major approaches,
emphasizing the similarities among frameworks and discussing potential future
directions.

Published
2023
Pages
8–29
Journal
IEEE SIGNAL PROCESSING MAGAZINE, vol. 40, no. 3, ISSN 1558-0792
DOI
10.1109/MSP.2023.3240008
UT WoS
000981974000003
EID Scopus
2-s2.0-85159861514
BibTeX 
@article{BUT185203,
  author="ŽMOLÍKOVÁ, K. and DELCROIX, M. and OCHIAI, T. and ČERNOCKÝ, J. and KINOSHITA, K. and YU, D.",
  title="Neural Target Speech Extraction: An overview",
  journal="IEEE SIGNAL PROCESSING MAGAZINE",
  year="2023",
  volume="40",
  number="3",
  pages="8--29",
  doi="10.1109/MSP.2023.3240008",
  issn="1558-0792",
  url="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10113382"
}
Files
Back to top