A Hybrid Parameterization Technique for Speaker Identification

Gómez Vilda, Pedro; Álvarez Marquina, Agustin; Mazaira Fernández, Luis Miguel; Fernández-Baillo Gallego de la Sacristana, Roberto; Nieto Lluis, Victor; Martínez Olalla, Rafael; Muñoz, Cristina y Rodellar Biarge, M. Victoria (2008). A Hybrid Parameterization Technique for Speaker Identification. En: "16th European Signal Processing Conference", 25/09/2008-29/09/2008, Lausanne, Switzerland.

Descripción

Título: A Hybrid Parameterization Technique for Speaker Identification
Autor/es:
  • Gómez Vilda, Pedro
  • Álvarez Marquina, Agustin
  • Mazaira Fernández, Luis Miguel
  • Fernández-Baillo Gallego de la Sacristana, Roberto
  • Nieto Lluis, Victor
  • Martínez Olalla, Rafael
  • Muñoz, Cristina
  • Rodellar Biarge, M. Victoria
Tipo de Documento: Ponencia en Congreso o Jornada (Artículo)
Título del Evento: 16th European Signal Processing Conference
Fechas del Evento: 25/09/2008-29/09/2008
Lugar del Evento: Lausanne, Switzerland
Título del Libro: Proceedings of the 16th European Signal Processing Conference (EUSIPCO 2008)
Fecha: 2008
Materias:
Escuela: Facultad de Informática (UPM) [antigua denominación]
Departamento: Arquitectura y Tecnología de Sistemas Informáticos
Licencias Creative Commons: Reconocimiento - Sin obra derivada - No comercial

Texto completo

[img]
Vista Previa
PDF (Document Portable Format) - Se necesita un visor de ficheros PDF, como GSview, Xpdf o Adobe Acrobat Reader
Descargar (868kB) | Vista Previa

Resumen

Classical parameterization techniques for Speaker Identification use the codification of the power spectral density of raw speech, not discriminating between articulatory features produced by vocal tract dynamics (acoustic-phonetics) from glottal source biometry. Through the present paper a study is conducted to separate voicing fragments of speech into vocal and glottal components, dominated respectively by the vocal tract transfer function estimated adaptively to track the acoustic-phonetic sequence of the message, and by the glottal characteristics of the speaker and the phonation gesture. The separation methodology is based in Joint Process Estimation under the un-correlation hypothesis between vocal and glottal spectral distributions. Its application on voiced speech is presented in the time and frequency domains. The parameterization methodology is also described. Speaker Identification experiments conducted on 245 speakers are shown comparing different parameterization strategies. The results confirm the better performance of decoupled parameterization compared against approaches based on plain speech parameterization.

Más información

ID de Registro: 4672
Identificador DC: http://oa.upm.es/4672/
Identificador OAI: oai:oa.upm.es:4672
URL Oficial: http://www.eurasip.org/Proceedings/Eusipco/Eusipco2008/program.html
Depositado por: Memoria Investigacion
Depositado el: 25 Oct 2010 11:35
Ultima Modificación: 20 Abr 2016 13:47
  • Open Access
  • Open Access
  • Sherpa-Romeo
    Compruebe si la revista anglosajona en la que ha publicado un artículo permite también su publicación en abierto.
  • Dulcinea
    Compruebe si la revista española en la que ha publicado un artículo permite también su publicación en abierto.
  • Recolecta
  • e-ciencia
  • Observatorio I+D+i UPM
  • OpenCourseWare UPM