Merging of Native and Non-native Speech for Low-resource Accented ASR - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

Merging of Native and Non-native Speech for Low-resource Accented ASR

Résumé

This paper presents our recent study on low-resource automatic speech recognition (ASR) system with accented speech. We propose multi-accent Subspace Gaussian Mixture Models (SGMM) and accent-specific Deep Neural Networks (DNN) for improving non-native ASR performance. In the SGMM framework, we present an original language weighting strategy to merge the globally shared parameters of two models based on native and non-native speech respectively. In the DNN framework, a native deep neural net is fine-tuned to non-native speech. Over the non-native baseline, we achieved relative improvement of 15 % for multi-accent SGMM and 34 % for accent-specific DNN with speaker adaptation.
Fichier principal
Vignette du fichier
SLSP2015-sarah.pdf (639.64 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01289140 , version 1 (29-01-2018)

Identifiants

Citer

Sarah Samson Juan, Laurent Besacier, Benjamin Lecouteux, Tien-Ping Tan. Merging of Native and Non-native Speech for Low-resource Accented ASR. 3rd International Conference on Statistical Language and Speech Processing, SLSP 2015, Nov 2015, Budapest, Hungary. ⟨10.1007/978-3-319-25789-1_24⟩. ⟨hal-01289140⟩
152 Consultations
237 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More