Skip navigation
Veuillez utiliser cette adresse pour citer ce document : http://repositorio.unb.br/handle/10482/10945
Fichier(s) constituant ce document :
Fichier Description TailleFormat 
ARTIGO_NoiseRobustSpeaker.pdf988,9 kBAdobe PDFVoir/Ouvrir
Titre: Noise-robust speaker recognition using reduced Multiconditional Gaussian Mixture models
Auteur(s): D’Almeida, Frederico Quadros
Nascimento, Francisco Assis de Oliveira
Berger, Pedro de Azevedo
Silva, Lúcio Martins da
Assunto:: Voz - ruído
Interação homem-máquina
Compressão de dados (Computação)
Reconhecimento automático da voz
Date de publication: 2008
Editeur: Brazilian Association of High Technology Experts (ABEAT)
Référence bibliographique: D'ALMEIDA, Frederico Quadros et al. Noise-robust speaker recognition using reducedMulticonditional gaussian mixture models. The International Journal of Forensic Computer Science, v. 3, n. 1, p. 60-69, 2008. Disponível em:<http://www.ijofcs.org/abstract-v03n1-pp06.html>. Acesso em: 19 jun. 2012. doi: 10.5769/J200801006
Résumé: Multiconditional Modeling is widely used to create noise-robust speaker recognition systems. However, the approach is computationally intensive. An alternative is to optimize the training condition set in order to achieve maximum noise robustness while using the smallest possible number of noise conditions during training. This paper establishes the optimal conditions for a noise-robust training model by considering audio material at different sampling rates and with different coding methods. Our results demonstrate that using approximately four training noise conditions is sufficient to guarantee robust models in the 60 dB to 10 dB Signal-to-Noise Ratio (SNR) range.
Licença:: Disponível sob Licença Creative Commons 3.0, que permite copiar, distribuir e transmitir o trabalho, desde que seja citado o autor e licenciante. Não permite o uso para fins comerciais nem a adaptação desta.
DOI: https://dx.doi.org/10.5769/J200801006
Collection(s) :Artigos publicados em periódicos e afins

Affichage détaillé " class="statisticsLink btn btn-primary" href="/jspui/handle/10482/10945/statistics">



Ce document est autorisé sous une licence de type Licence Creative Commons Creative Commons