Extending AuToBI to prominence detection in European Portuguese
Autor(a) principal: | |
---|---|
Data de Publicação: | 2014 |
Outros Autores: | , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10451/31086 |
Resumo: | This paper describes our exploratory work in applying the Automatic ToBI annotation system (AuToBI), originally developed for Standard American English, to European Portuguese. This work is motivated by the current availability of large amounts of (highly spontaneous) transcribed data and the need to further enrich those transcripts with prosodic information. Manual prosodic annotation, however, is almost impractical for extensive data sets. For that reason, automatic systems such as AuToBi stand as an alternate solution. We have started by applying the AuToBI prosodic event detection system using the existing English models to the prediction of prominent prosodic events (accents) in European Portuguese. This approach achieved an overall accuracy of 74% for prominence detection, similar to state-of-the-art results for other languages. Later, we have trained new models using prepared and spontaneous Portuguese data, achieving a considerable improvement of about 6% accuracy (absolute) over the existing English models. The achieved results are quite encouraging and provide a starting point for automatically predicting prominent events in European Portuguese. |
id |
RCAP_74757bcb38302e34dcced90aaae0c70a |
---|---|
oai_identifier_str |
oai:repositorio.ul.pt:10451/31086 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Extending AuToBI to prominence detection in European PortugueseProsodyAutomatic prosodic labeling systemSpontaneous speechThis paper describes our exploratory work in applying the Automatic ToBI annotation system (AuToBI), originally developed for Standard American English, to European Portuguese. This work is motivated by the current availability of large amounts of (highly spontaneous) transcribed data and the need to further enrich those transcripts with prosodic information. Manual prosodic annotation, however, is almost impractical for extensive data sets. For that reason, automatic systems such as AuToBi stand as an alternate solution. We have started by applying the AuToBI prosodic event detection system using the existing English models to the prediction of prominent prosodic events (accents) in European Portuguese. This approach achieved an overall accuracy of 74% for prominence detection, similar to state-of-the-art results for other languages. Later, we have trained new models using prepared and spontaneous Portuguese data, achieving a considerable improvement of about 6% accuracy (absolute) over the existing English models. The achieved results are quite encouraging and provide a starting point for automatically predicting prominent events in European Portuguese.Urbana, ILRepositório da Universidade de LisboaMoniz, HelenaMata, Ana IsabelHirschberg, JuliaBatista, FernandoRosenberg, AndrewTrancoso, Isabel2018-01-28T15:26:11Z20142014-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttp://hdl.handle.net/10451/31086engMoniz, H., Mata, A. I., Hirschberg, J. & Batista, F., Rosenberg, A. & Trancoso, I. (2014) Extending AuToBI to Prominence Detection in European Portuguese, in Speech Prosody.info:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-11-08T16:24:15Zoai:repositorio.ul.pt:10451/31086Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T21:46:36.804135Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Extending AuToBI to prominence detection in European Portuguese |
title |
Extending AuToBI to prominence detection in European Portuguese |
spellingShingle |
Extending AuToBI to prominence detection in European Portuguese Moniz, Helena Prosody Automatic prosodic labeling system Spontaneous speech |
title_short |
Extending AuToBI to prominence detection in European Portuguese |
title_full |
Extending AuToBI to prominence detection in European Portuguese |
title_fullStr |
Extending AuToBI to prominence detection in European Portuguese |
title_full_unstemmed |
Extending AuToBI to prominence detection in European Portuguese |
title_sort |
Extending AuToBI to prominence detection in European Portuguese |
author |
Moniz, Helena |
author_facet |
Moniz, Helena Mata, Ana Isabel Hirschberg, Julia Batista, Fernando Rosenberg, Andrew Trancoso, Isabel |
author_role |
author |
author2 |
Mata, Ana Isabel Hirschberg, Julia Batista, Fernando Rosenberg, Andrew Trancoso, Isabel |
author2_role |
author author author author author |
dc.contributor.none.fl_str_mv |
Repositório da Universidade de Lisboa |
dc.contributor.author.fl_str_mv |
Moniz, Helena Mata, Ana Isabel Hirschberg, Julia Batista, Fernando Rosenberg, Andrew Trancoso, Isabel |
dc.subject.por.fl_str_mv |
Prosody Automatic prosodic labeling system Spontaneous speech |
topic |
Prosody Automatic prosodic labeling system Spontaneous speech |
description |
This paper describes our exploratory work in applying the Automatic ToBI annotation system (AuToBI), originally developed for Standard American English, to European Portuguese. This work is motivated by the current availability of large amounts of (highly spontaneous) transcribed data and the need to further enrich those transcripts with prosodic information. Manual prosodic annotation, however, is almost impractical for extensive data sets. For that reason, automatic systems such as AuToBi stand as an alternate solution. We have started by applying the AuToBI prosodic event detection system using the existing English models to the prediction of prominent prosodic events (accents) in European Portuguese. This approach achieved an overall accuracy of 74% for prominence detection, similar to state-of-the-art results for other languages. Later, we have trained new models using prepared and spontaneous Portuguese data, achieving a considerable improvement of about 6% accuracy (absolute) over the existing English models. The achieved results are quite encouraging and provide a starting point for automatically predicting prominent events in European Portuguese. |
publishDate |
2014 |
dc.date.none.fl_str_mv |
2014 2014-01-01T00:00:00Z 2018-01-28T15:26:11Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10451/31086 |
url |
http://hdl.handle.net/10451/31086 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Moniz, H., Mata, A. I., Hirschberg, J. & Batista, F., Rosenberg, A. & Trancoso, I. (2014) Extending AuToBI to Prominence Detection in European Portuguese, in Speech Prosody. |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Urbana, IL |
publisher.none.fl_str_mv |
Urbana, IL |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799134391169974272 |