Evaluation of PNCC and extended spectral subtraction methods for robust speech recognition - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

Evaluation of PNCC and extended spectral subtraction methods for robust speech recognition

Résumé

This paper evaluates the robustness of different approaches for speech recognition with respect to signal-to-noise ratio (SNR), to signal level and to presence of non-speech data before and after utterances to be recognized. Three types of noise robust features are considered: Power Normalized Cepstral Coefficients (PNCC), Mel-Frequency Cepstral Coefficients (MFCC) after applying an extended spectral subtraction method, and Sphinx embedded denoising features from recent sphinx versions. Although removing C0 in MFCC-based features leads to a slight decrease in speech recognition performance, it makes the speech recognition system independent on the speech signal level. With multi-condition training, the three sets of noise-robust features lead to a rather similar behavior of performance with respect to SNR and presence of non-speech data. Overall, best performance is achieved with the extended spectral subtraction approach. Also, the performance of the PNCC features appears to be dependent on the initialization of the normalization factor.
Fichier principal
Vignette du fichier
Eusipco-FinalUpdate-June2015--1570104069.pdf (308.57 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01183645 , version 1 (10-08-2015)

Identifiants

  • HAL Id : hal-01183645 , version 1

Citer

Thibaut Fux, Denis Jouvet. Evaluation of PNCC and extended spectral subtraction methods for robust speech recognition. EUSIPCO 2015 - 23rd European Signal Processing Conference , Aug 2015, Nice, France. ⟨hal-01183645⟩
211 Consultations
608 Téléchargements

Partager

Gmail Facebook X LinkedIn More