Combining EigenVoices and Structural MLLR for Speaker Adaptation - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2003

Combining EigenVoices and Structural MLLR for Speaker Adaptation

Irina Illina
Dominique Fohr

Résumé

This papers considers the problem of speaker adaptation of acoustic models in speech recognition. We have investigated four possible methods which integrate the concepts of both Structural Maximum Likelihood Linear regression (SMLLR) and EigenVoices-based technique (EV) to adapt the Gaussian means of the speaker independant models for a new speaker. The experiments were evaluated using the speech recognition engine ESPERE on the data of the corpus Resource Management. They show that all of the proposed methods can improve the performances of an automatic speech recognition system (ASRS) in supervised batch adaptation as efficiently as SMLLR and EigenVoices-based techniques whatever the amount of adaptation data is available. For an unsupervised incremental adaptation, only the approach SMLLR+SEV gives the best results.

Domaines

Autre [cs.OH]
Fichier non déposé

Dates et versions

inria-00099746 , version 1 (26-09-2006)

Identifiants

  • HAL Id : inria-00099746 , version 1

Citer

Fabrice Lauri, Irina Illina, Dominique Fohr. Combining EigenVoices and Structural MLLR for Speaker Adaptation. IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP'03, Apr 2003, Hong Kong, China, 4 p. ⟨inria-00099746⟩
105 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More