The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Cet article présente une nouvelle méthode d'extraction de caractéristiques robuste au bruit pour la reconnaissance vocale. Il est basé sur la rendre la méthode d'estimation du spectre de puissance MVDR (Minimum Variance Distortionless Response) robuste au bruit. Cette robustesse est obtenue en modifiant la contrainte sans distorsion de la méthode d'estimation spectrale MVDR via une pondération des valeurs du spectre de puissance des sous-bandes en fonction des rapports signal sur bruit des sous-bandes. La pondération optimale est obtenue en utilisant les résultats expérimentaux de la psychoacoustique. D'après nos expériences, cette technique parvient à modifier le spectre de puissance des signaux vocaux et à le rendre robuste au bruit. La méthode ci-dessus, lorsqu'elle a été évaluée sur la tâche Aurora 2 à des fins de reconnaissance, a surpassé à la fois les fonctionnalités MFCC comme référence et les fonctionnalités basées sur MVDR dans différentes conditions bruyantes.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Sanaz SEYEDIN, Seyed Mohammad AHADI, "A New Subband-Weighted MVDR-Based Front-End for Robust Speech Recognition" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 8, pp. 2252-2261, August 2010, doi: 10.1587/transinf.E93.D.2252.
Abstract: This paper presents a novel noise-robust feature extraction method for speech recognition. It is based on making the Minimum Variance Distortionless Response (MVDR) power spectrum estimation method robust against noise. This robustness is obtained by modifying the distortionless constraint of the MVDR spectral estimation method via weighting the sub-band power spectrum values based on the sub-band signal to noise ratios. The optimum weighting is obtained by employing the experimental findings of psychoacoustics. According to our experiments, this technique is successful in modifying the power spectrum of speech signals and making it robust against noise. The above method, when evaluated on Aurora 2 task for recognition purposes, outperformed both the MFCC features as the baseline and the MVDR-based features in different noisy conditions.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.2252/_p
Copier
@ARTICLE{e93-d_8_2252,
author={Sanaz SEYEDIN, Seyed Mohammad AHADI, },
journal={IEICE TRANSACTIONS on Information},
title={A New Subband-Weighted MVDR-Based Front-End for Robust Speech Recognition},
year={2010},
volume={E93-D},
number={8},
pages={2252-2261},
abstract={This paper presents a novel noise-robust feature extraction method for speech recognition. It is based on making the Minimum Variance Distortionless Response (MVDR) power spectrum estimation method robust against noise. This robustness is obtained by modifying the distortionless constraint of the MVDR spectral estimation method via weighting the sub-band power spectrum values based on the sub-band signal to noise ratios. The optimum weighting is obtained by employing the experimental findings of psychoacoustics. According to our experiments, this technique is successful in modifying the power spectrum of speech signals and making it robust against noise. The above method, when evaluated on Aurora 2 task for recognition purposes, outperformed both the MFCC features as the baseline and the MVDR-based features in different noisy conditions.},
keywords={},
doi={10.1587/transinf.E93.D.2252},
ISSN={1745-1361},
month={August},}
Copier
TY - JOUR
TI - A New Subband-Weighted MVDR-Based Front-End for Robust Speech Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 2252
EP - 2261
AU - Sanaz SEYEDIN
AU - Seyed Mohammad AHADI
PY - 2010
DO - 10.1587/transinf.E93.D.2252
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2010
AB - This paper presents a novel noise-robust feature extraction method for speech recognition. It is based on making the Minimum Variance Distortionless Response (MVDR) power spectrum estimation method robust against noise. This robustness is obtained by modifying the distortionless constraint of the MVDR spectral estimation method via weighting the sub-band power spectrum values based on the sub-band signal to noise ratios. The optimum weighting is obtained by employing the experimental findings of psychoacoustics. According to our experiments, this technique is successful in modifying the power spectrum of speech signals and making it robust against noise. The above method, when evaluated on Aurora 2 task for recognition purposes, outperformed both the MFCC features as the baseline and the MVDR-based features in different noisy conditions.
ER -