Back
  1. Home
  3. Detailled Reference
Download
Paper published in a book (Scientific congresses, symposiums and conference proceedings)
LUX-ASR: Building an ASR system for the Luxembourgish language
GILLES, Peter; HOSSEINI KIVANANI, Nina; HILLAH, Léopold Edem Ayité
2023 • In IEEE, Spoken Language Technology (Ed.) Proceedings - 2022 IEEE Spoken Language Technology Workshop (SLT)
Peer reviewed
Permalink
https://hdl.handle.net/10993/55105
 

Files (1)Send toDetailsStatisticsBibliographySimilar publications

Files

Full Text
0001147.pdf
Author preprint (168.25 kB)
Download

All documents in ORBilu are protected by a user license.

Send to


Details


Keywords :
Automatic Speech Recognition; Luxembourgish
Abstract :
[en] We present a first system for automatic speech recognition (ASR) for the low-resource language Luxembourgish. By applying transfer-learning, we were able to fine-tune Meta’s wav2vec2-xls-r-300m checkpoint with 35 hours of labeled Luxembourgish speech data. The best word error rate received lies at 14.47.
Disciplines :
Computer science
Author, co-author :
GILLES, Peter  ;  University of Luxembourg > Faculty of Humanities, Education and Social Sciences (FHSE) > Department of Humanities (DHUM)
HOSSEINI KIVANANI, Nina  ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
HILLAH, Léopold Edem Ayité  ;  University of Luxembourg > Faculty of Science, Technology and Medecine (FSTM)
External co-authors :
no
Language :
English
Title :
LUX-ASR: Building an ASR system for the Luxembourgish language
Publication date :
2023
Event name :
2022 IEEE Spoken Language Technology Workshop (SLT) SLT 2022
Event organizer :
SLT
Event place :
Doha, Qatar
Event date :
from 09-01-2023 to 12-01-2023
Audience :
International
Main work title :
Proceedings - 2022 IEEE Spoken Language Technology Workshop (SLT)
Editor :
IEEE, Spoken Language Technology
ISBN/EAN :
978-1-6654-7189-3
Pages :
1147-1149
Peer reviewed :
Peer reviewed
Focus Area :
Computational Sciences
Available on ORBilu :
since 09 May 2023

Statistics

Number of views
231 (13 by Unilu)
Number of downloads
184 (18 by Unilu)
More statistics

Bibliography

Similar publications


Contact ORBilu