Mostrar el registro sencillo del ítem

dc.contributor.author
Ferrer, Luciana  
dc.contributor.author
Castan, Diego  
dc.contributor.author
Mclaren, Mitchell  
dc.contributor.author
Lawson, Aaron  
dc.date.available
2023-07-20T13:53:03Z  
dc.date.issued
2022-07  
dc.identifier.citation
Ferrer, Luciana; Castan, Diego; Mclaren, Mitchell; Lawson, Aaron; A Discriminative Hierarchical PLDA-Based Model for Spoken Language Recognition; Institute of Electrical and Electronics Engineers; IEEE/ACM Transactions on Audio Speech and Language Processing; 30; 7-2022; 2396-2410  
dc.identifier.issn
2329-9304  
dc.identifier.uri
http://hdl.handle.net/11336/204631  
dc.description.abstract
Spoken language recognition (SLR) refers to the automatic process used to determine the language present in a speech sample. SLR is an important task in its own right, for example, as a tool to analyze or categorize large amounts of multi-lingual data. Further, it is also an essential tool for selecting downstream applications in a work flow, for example, to chose appropriate speech recognition or machine translation models. SLR systems are usually composed of two stages, one where an embedding representing the audio sample is extracted and a second one which computes the final scores for each language. In this work, we approach the SLR task as a detection problem and implement the second stage as a probabilistic linear discriminant analysis (PLDA) model. We show that discriminative training of the PLDA parameters gives large gains with respect to the usual generative training. Further, we propose a novel hierarchical approach where two PLDA models are trained, one to generate scores for clusters of highly-related languages and a second one to generate scores conditional to each cluster. The final language detection scores are computed as a combination of these two sets of scores. The complete model is trained discriminatively to optimize a cross-entropy objective. We show that this hierarchical approach consistently outperforms the non-hierarchical one for detection of highly related languages, in many cases by large margins. We train our systems on a collection of datasets including over 100 languages, and test them both on matched and mismatched conditions, showing that the gains are robust to condition mismatch.  
dc.format
application/pdf  
dc.language.iso
eng  
dc.publisher
Institute of Electrical and Electronics Engineers  
dc.rights
info:eu-repo/semantics/restrictedAccess  
dc.rights.uri
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/  
dc.subject
DISCRIMINATIVE TRAINING  
dc.subject
PROBABILISTIC LINEAR DISCRIMINANT ANALYSIS  
dc.subject
SPOKEN LANGUAGE RECOGNITION  
dc.subject.classification
Ciencias de la Información y Bioinformática  
dc.subject.classification
Ciencias de la Computación e Información  
dc.subject.classification
CIENCIAS NATURALES Y EXACTAS  
dc.title
A Discriminative Hierarchical PLDA-Based Model for Spoken Language Recognition  
dc.type
info:eu-repo/semantics/article  
dc.type
info:ar-repo/semantics/artículo  
dc.type
info:eu-repo/semantics/publishedVersion  
dc.date.updated
2023-07-07T22:23:43Z  
dc.journal.volume
30  
dc.journal.pagination
2396-2410  
dc.journal.pais
Estados Unidos  
dc.journal.ciudad
Nueva Jersey  
dc.description.fil
Fil: Ferrer, Luciana. Consejo Nacional de Investigaciones Científicas y Técnicas. Oficina de Coordinación Administrativa Ciudad Universitaria. Instituto de Investigación en Ciencias de la Computación. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales. Instituto de Investigación en Ciencias de la Computación; Argentina  
dc.description.fil
Fil: Castan, Diego. Sri International; Estados Unidos  
dc.description.fil
Fil: Mclaren, Mitchell. Sri International; Estados Unidos  
dc.description.fil
Fil: Lawson, Aaron. Sri International; Estados Unidos  
dc.journal.title
IEEE/ACM Transactions on Audio Speech and Language Processing  
dc.relation.alternativeid
info:eu-repo/semantics/altIdentifier/doi/http://dx.doi.org/10.1109/TASLP.2022.3190736  
dc.relation.alternativeid
info:eu-repo/semantics/altIdentifier/url/https://ieeexplore.ieee.org/document/9844653