Show simple item record

dc.creatorSaz, Oscar 
dc.creatorLleida, Eduardo 
dc.creatorRodríguez-Dueñas, William R. 
dc.date.accessioned2020-08-28T15:47:55Z
dc.date.available2020-08-28T15:47:55Z
dc.date.created2009-11
dc.identifier.issnISBN: 978-1-60558-690-8
dc.identifier.urihttps://repository.urosario.edu.co/handle/10336/28304
dc.description.abstractThis paper deals with the problematic of speaker variability in a task of pronunciation verification for the speech therapy of children and young adults in Computer-Aided Pronunciation Training (CAPT) tools. The baseline system evaluates two different score normalization techniques: Traditional Test normalization (T-norm), and a novel Nbest based normalization that outperforms the first by normalizing to the log-likelihood score of the first alternative phoneme in an unconstrained N-best list. When performing speaker adaptation, the use of all the adaptation data from the speaker improves the performance measured in Equal Error Rate (EER) of these systems compared to the speaker independent systems; but this can be outperformed by more precise models that only adapt to the correctly pronounced phonetic units as labeled by a set of human experts. The best EER obtained in all experiments is 15.63% when using both elements: Score normalization and speaker adaptation. The possibility of automatizing a more precise adaptation without the human intervention is finally proposed and discussed.
dc.format.mimetypeapplication/pdf
dc.language.isoeng
dc.relation.ispartofWOCCI '09: Proceedings of the 2nd Workshop on Child, Computer and Interaction
dc.relation.ispartofCMI-MLMI '09: International Conference On Multimodal Interfaces/Workshop On Machine Learning For Multimodal Interfaces Cambridge Massachusetts, ISBN: 978-1-60558-690-89 (2009); pp. 1-5
dc.relation.urihttps://dl.acm.org/doi/10.1145/1640377.1640388
dc.sourceWOCCI '09: Proceedings of the 2nd Workshop on Child, Computer and Interaction
dc.sourceCMI-MLMI '09: International Conference On Multimodal Interfaces/Workshop On Machine Learning For Multimodal Interfaces Cambridge Massachusetts (November, 2009)
dc.titleAvoiding speaker variability in pronunciation verification of children's disordered speech
dc.typearticle
dc.publisherAssociation for Computing Machinery
dc.subject.keywordPronunciation evaluation
dc.subject.keywordChildren speech
dc.subject.keywordSpeech disorders
dc.rights.accesRightsinfo:eu-repo/semantics/restrictedAccess
dc.type.spaArtículo
dc.rights.accesoRestringido (Acceso a grupos específicos)
dc.type.hasVersioninfo:eu-repo/semantics/publishedVersion
dc.identifier.doihttps://doi.org/10.1145/1640377.1640388
dc.title.TranslatedTitleEvitar la variabilidad del hablante en la verificación de la pronunciación del habla desordenada de los niños
dc.relation.citationTitleWOCCI `09: Proceedings of the 2nd Workshop on Child, Computer and Interaction;CMI-MLMI `09: International Conference On Multimodal Interfaces/Workshop On Machine Learning For Multimodal Interfaces Cambridge Massachusetts (November, 2009)


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record