Multitask Detection of Speaker Changes, Overlapping Speech and Voice Activity Using Wav2vec 2.0

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F23%3A43968753" target="_blank" >RIV/49777513:23520/23:43968753 - isvavai.cz</a>
Result on the web
<a href="https://ieeexplore.ieee.org/document/10094972/" target="_blank" >https://ieeexplore.ieee.org/document/10094972/</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/ICASSP49357.2023.10094972" target="_blank" >10.1109/ICASSP49357.2023.10094972</a>

Result language
angličtina
Original language name
Multitask Detection of Speaker Changes, Overlapping Speech and Voice Activity Using Wav2vec 2.0
Original language description
Self-supervised learning approaches have lately achieved great success on a broad spectrum of machine learning problems. In the field of speech processing, one of the most successful recent self-supervised models is wav2vec 2.0. In this paper, we explore the effectiveness of this model on three basic speech classification tasks: speaker change detection, overlapped speech detection, and voice activity detection. First, we concentrate on only one task – speaker change detection – where our proposed system surpasses the previously reported results on four different corpora, and achieves comparable performance even when trained on out-of-domain data from an artificially designed dataset. Then we expand our approach to tackle all three tasks in a single multitask system with state-of-the-art performance on the AMI corpus. The implementation of the algorithms in this paper is publicly available at https://github.com/mkunes/w2v2_audioFrameClassification.
Czech name
—
Czech description
—

Project
Result was created during the realization of more than one project. More information in the Projects tab.
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2023
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
ISBN
978-1-72816-327-7
ISSN
1520-6149
e-ISSN
2379-190X
Number of pages
5
Pages from-to
"nestrankovano"
Publisher name
IEEE
Place of publication
New York
Event location
Rhodes Island, Greece
Event date
Jun 4, 2023
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)