- -

Automatic speech recognition in cocktail-party situations : a specific training for separated speech

RiuNet: Institutional repository of the Polithecnic University of Valencia

Share/Send to

Cited by

Statistics

Automatic speech recognition in cocktail-party situations : a specific training for separated speech

Show simple item record

Files in this item

dc.contributor.author Martí Guerola, Amparo es_ES
dc.contributor.author Cobos Serrano, Máximo es_ES
dc.contributor.author López Monfort, José Javier es_ES
dc.date.accessioned 2015-11-02T08:28:06Z
dc.date.issued 2012-02
dc.identifier.issn 0001-4966
dc.identifier.uri http://hdl.handle.net/10251/56829
dc.description.abstract Automatic speech recognition (ASR) refers to the task of extracting a transcription of the linguistic content of an acoustical speech signal automatically. Despite several decades of research in this important area of acoustic signal processing, the accuracy of ASR systems is still far behind human performance, especially in adverse acoustic scenarios. In this context, one of the most challenging situations is the one concerning simultaneous speech in cocktail-party environments. Although source separation methods have already been investigated to deal with this problem, the separation process is not perfect and the resulting artifacts pose an additional problem to ASR performance. In this paper, a specific training to improve the percentage of recognized words in real simultaneous speech cases is proposed. The combination of source separation and this specific training is explored and evaluated under different acoustical conditions, leading to improvements of up to a 35% in ASR performance. (C) 2012 Acoustical Society of America. [DOI: 10.1121/1.3675001] es_ES
dc.description.sponsorship The Spanish Ministry of Science and Innovation supported this work under Grant No. TEC2009-14414-C03-01. en_EN
dc.language Inglés es_ES
dc.publisher Acoustical Society of America es_ES
dc.relation Spanish Ministry of Science and Innovation [TEC2009-14414-C03-01] es_ES
dc.relation.ispartof Journal of the Acoustical Society of America es_ES
dc.rights Reserva de todos los derechos es_ES
dc.subject Automatic speech recognition es_ES
dc.subject Human performance es_ES
dc.subject Separation process es_ES
dc.subject Source separation es_ES
dc.subject Speech signals es_ES
dc.subject Acoustics es_ES
dc.subject Physics es_ES
dc.subject Separation es_ES
dc.subject Algorithm es_ES
dc.subject Article es_ES
dc.subject Human es_ES
dc.subject Noise es_ES
dc.subject Perception es_ES
dc.subject Speech es_ES
dc.subject Speech perception es_ES
dc.subject Standard es_ES
dc.subject Algorithms es_ES
dc.subject Humans es_ES
dc.subject Perceptual Masking es_ES
dc.subject.classification TEORIA DE LA SEÑAL Y COMUNICACIONES es_ES
dc.title Automatic speech recognition in cocktail-party situations : a specific training for separated speech es_ES
dc.type Artículo es_ES
dc.embargo.lift 10000-01-01
dc.embargo.terms forever es_ES
dc.identifier.doi 10.1121/1.3675001
dc.rights.accessRights Cerrado es_ES
dc.contributor.affiliation Universitat Politècnica de València. Departamento de Comunicaciones - Departament de Comunicacions es_ES
dc.contributor.affiliation Universitat Politècnica de València. Instituto Universitario de Telecomunicación y Aplicaciones Multimedia - Institut Universitari de Telecomunicacions i Aplicacions Multimèdia es_ES
dc.description.bibliographicCitation Martí Guerola, A.; Cobos Serrano, M.; López Monfort, JJ. (2012). Automatic speech recognition in cocktail-party situations : a specific training for separated speech. Journal of the Acoustical Society of America. 131(2):1529-1535. doi:10.1121/1.3675001 es_ES
dc.description.accrualMethod Senia es_ES
dc.relation.publisherversion http://dx.doi.org/10.1121/1.3675001 es_ES
dc.description.upvformatpinicio 1529 es_ES
dc.description.upvformatpfin 1535 es_ES
dc.type.version info:eu-repo/semantics/publishedVersion es_ES
dc.description.volume 131 es_ES
dc.description.issue 2 es_ES
dc.relation.senia 211996 es_ES
dc.identifier.eissn 1520-8524


This item appears in the following Collection(s)

Show simple item record