Multilingual Automatic Speech Recognition for Scandinavian Languages

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F23%3AZREZPYUQ" target="_blank" >RIV/00216208:11320/23:ZREZPYUQ - isvavai.cz</a>

Result on the web

<a href="https://aclanthology.org/2023.nodalida-1.46/" target="_blank" >https://aclanthology.org/2023.nodalida-1.46/</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Multilingual Automatic Speech Recognition for Scandinavian Languages

Original language description

"We investigate the effectiveness of multilingual automatic speech recognition models for Scandinavian languages by further fine-tuning a Swedish model on Swedish, Danish, and Norwegian. We first explore zero-shot models, which perform poorly across the three languages. However, we show that a multilingual model based on a strong Swedish model, further fine-tuned on all three languages, performs well for Norwegian and Danish, with a relatively low decrease in the performance for Swedish. With a language classification module, we improve the performance of the multilingual model even further."

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

—

Continuities

—

Publication year

2023

Confidentiality

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

Article name in the collection

"Proceedings of the 24th Nordic Conference on Computational Linguistics (NoDaLiDa)"

ISBN

978-99-1621-999-7

ISSN

1736-8197

e-ISSN

—

Number of pages

7

Pages from-to

460-466

Publisher name

arXiv

Place of publication

—

Event location

Singapore

Event date

Jan 1, 2023

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—