Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR

In this paper, a semisupervised speech data extraction method is presented and applied to create a new dataset designed for the development of fully bilingual Automatic Speech Recognition (ASR) systems for Basque and Spanish. The dataset is drawn from an extensive collection of Basque Parliament ple...

Descripción completa

Detalles Bibliográficos
Autores: Peñagarikano Badiola, Mikel, Varona Fernández, Amparo, Bordel García, Germán, Rodríguez Fuentes, Luis Javier
Tipo de recurso: artículo
Fecha de publicación:2023
País:España
Institución:Universidad del País Vasco
Repositorio:Addi. Archivo Digital para la Docencia y la Investigación
OAI Identifier:oai:addi.ehu.eus:10810/62077
Acceso en línea:http://hdl.handle.net/10810/62077
Access Level:acceso abierto
Palabra clave:automatic speech recognition
multilingual speech
low-resource languages
code switching
semisupervised learning
spoken language resources
id ES_ba4830aee2f8124fd7c8b3a91c920d80
oai_identifier_str oai:addi.ehu.eus:10810/62077
network_acronym_str ES
network_name_str España
repository_id_str
spelling Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASRPeñagarikano Badiola, MikelVarona Fernández, AmparoBordel García, GermánRodríguez Fuentes, Luis Javierautomatic speech recognitionmultilingual speechlow-resource languagescode switchingsemisupervised learningspoken language resourcesIn this paper, a semisupervised speech data extraction method is presented and applied to create a new dataset designed for the development of fully bilingual Automatic Speech Recognition (ASR) systems for Basque and Spanish. The dataset is drawn from an extensive collection of Basque Parliament plenary sessions containing frequent code switchings. Since session minutes are not exact, only the most reliable speech segments are kept for training. To that end, we use phonetic similarity scores between nominal and recognized phone sequences. The process starts with baseline acoustic models trained on generic out-of-domain data, then iteratively updates the models with the extracted data and applies the updated models to refine the training dataset until the observed improvement between two iterations becomes small enough. A development dataset, involving five plenary sessions not used for training, has been manually audited for tuning and evaluation purposes. Cross-validation experiments (with 20 random partitions) have been carried out on the development dataset, using the baseline and the iteratively updated models. On average, Word Error Rate (WER) reduces from 16.57% (baseline) to 4.41% (first iteration) and further to 4.02% (second iteration), which corresponds to relative WER reductions of 73.4% and 8.8%, respectively. When considering only Basque segments, WER reduces on average from 16.57% (baseline) to 5.51% (first iteration) and further to 5.13% (second iteration), which corresponds to relative WER reductions of 66.7% and 6.9%, respectively. As a result of this work, a new bilingual Basque–Spanish resource has been produced based on Basque Parliament sessions, including 998 h of training data (audio segments + transcriptions), a development set (17 h long) designed for tuning and evaluation under a cross-validation scheme and a fully bilingual trigram language model.This work was partially funded by the Spanish Ministry of Science and Innovation (OPEN-SPEECH project, PID2019-106424RB-I00) and by the Basque Government under the general support program to research groups (IT-1704-22).MDPI2023202320232023info:eu-repo/semantics/articleapplication/pdfhttp://hdl.handle.net/10810/62077reponame:Addi. Archivo Digital para la Docencia y la Investigacióninstname:Universidad del País VascoInglésinfo:eu-repo/grantAgreement/MICINN/PID2019-106424RB-I00/https://www.mdpi.com/2076-3417/13/14/8492info:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by/4.0/© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/ 4.0/).oai:addi.ehu.eus:10810/620772026-06-18T09:23:17Z
dc.title.none.fl_str_mv Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
title Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
spellingShingle Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
Peñagarikano Badiola, Mikel
automatic speech recognition
multilingual speech
low-resource languages
code switching
semisupervised learning
spoken language resources
title_short Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
title_full Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
title_fullStr Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
title_full_unstemmed Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
title_sort Semisupervised Speech Data Extraction from Basque Parliament Sessions and Validation on Fully Bilingual Basque–Spanish ASR
dc.creator.none.fl_str_mv Peñagarikano Badiola, Mikel
Varona Fernández, Amparo
Bordel García, Germán
Rodríguez Fuentes, Luis Javier
author Peñagarikano Badiola, Mikel
author_facet Peñagarikano Badiola, Mikel
Varona Fernández, Amparo
Bordel García, Germán
Rodríguez Fuentes, Luis Javier
author_role author
author2 Varona Fernández, Amparo
Bordel García, Germán
Rodríguez Fuentes, Luis Javier
author2_role author
author
author
dc.subject.none.fl_str_mv automatic speech recognition
multilingual speech
low-resource languages
code switching
semisupervised learning
spoken language resources
topic automatic speech recognition
multilingual speech
low-resource languages
code switching
semisupervised learning
spoken language resources
description In this paper, a semisupervised speech data extraction method is presented and applied to create a new dataset designed for the development of fully bilingual Automatic Speech Recognition (ASR) systems for Basque and Spanish. The dataset is drawn from an extensive collection of Basque Parliament plenary sessions containing frequent code switchings. Since session minutes are not exact, only the most reliable speech segments are kept for training. To that end, we use phonetic similarity scores between nominal and recognized phone sequences. The process starts with baseline acoustic models trained on generic out-of-domain data, then iteratively updates the models with the extracted data and applies the updated models to refine the training dataset until the observed improvement between two iterations becomes small enough. A development dataset, involving five plenary sessions not used for training, has been manually audited for tuning and evaluation purposes. Cross-validation experiments (with 20 random partitions) have been carried out on the development dataset, using the baseline and the iteratively updated models. On average, Word Error Rate (WER) reduces from 16.57% (baseline) to 4.41% (first iteration) and further to 4.02% (second iteration), which corresponds to relative WER reductions of 73.4% and 8.8%, respectively. When considering only Basque segments, WER reduces on average from 16.57% (baseline) to 5.51% (first iteration) and further to 5.13% (second iteration), which corresponds to relative WER reductions of 66.7% and 6.9%, respectively. As a result of this work, a new bilingual Basque–Spanish resource has been produced based on Basque Parliament sessions, including 998 h of training data (audio segments + transcriptions), a development set (17 h long) designed for tuning and evaluation under a cross-validation scheme and a fully bilingual trigram language model.
publishDate 2023
dc.date.none.fl_str_mv 2023
2023
2023
2023
dc.type.none.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv http://hdl.handle.net/10810/62077
url http://hdl.handle.net/10810/62077
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv info:eu-repo/grantAgreement/MICINN/PID2019-106424RB-I00/
https://www.mdpi.com/2076-3417/13/14/8492
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by/4.0/
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv MDPI
publisher.none.fl_str_mv MDPI
dc.source.none.fl_str_mv reponame:Addi. Archivo Digital para la Docencia y la Investigación
instname:Universidad del País Vasco
instname_str Universidad del País Vasco
reponame_str Addi. Archivo Digital para la Docencia y la Investigación
collection Addi. Archivo Digital para la Docencia y la Investigación
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869417865050849280
score 15,300724

Similares en LA Referencia