Design choices for PixIT-based speaker-attributed ASR: Team ToTaTo at the NOTSOFAR-1 challenge
author
Kalda, Joonas
Baroudi, Séverin
Lebourdais, Martin
Pages, Clement
Marxer, Ricard
Alumäe, Tanel
Bredin, Hervé
statement of authorship
Joonas Kalda, Séverin Baroudi, Martin Lebourdais, Clément Pagés, Ricard Marxer, Tanel Alumäe, Hervé Bredin
source
Computer Speech & Language
publisher
Elsevier
journal volume number month
vol. 95
year of publication
2026
pages
art. 101824, 16 p. : ill
url
https://doi.org/10.1016/j.csl.2025.101824
subject term
kõnetuvastus
kõne
keyword
Speaker diarization
Speaker-attributed automatic speech
recognition
Speaker embeddings
SSL models
Joint training
ISSN
0885-2308
1095-8363
notes
Bibliogr. p. 15-16
Open Access
Open Access
scientific publication
teaduspublikatsioon
classifier
1.1
TalTech department
tarkvarateaduse instituut
language
inglise