Probabilistic Speaker Pronunciation Adaptation for Spontaneous Speech Synthesis Using Linguistic Features - IMT - Institut Mines-Télécom Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

Probabilistic Speaker Pronunciation Adaptation for Spontaneous Speech Synthesis Using Linguistic Features

Résumé

Pronunciation adaptation consists in predicting pronunciation variants of words and utterances based on their standard pronunciation and a target style. This is a key issue in text-to-speech as those variants bring expressiveness to synthetic speech, especially when considering a spontaneous style. This paper presents a new pronunciation adaptation method which adapts standard pronunciations to the style of individual speakers in a context of spontaneous speech. Its originality and strength are to solely rely on linguistic features and to consider a probabilistic machine learning framework, namely conditional random fields, to produce the adapted pronunciations. Features are first selected in a series of experiments, then combined to produce the final adaptation method. Backend experiments on the Buckeye conversational English speech corpus show that adapted pronunciations significantly better reflect spontaneous speech than standard ones, and that even better could be achieved if considering alternative predictions.
Fichier principal
Vignette du fichier
94490238.pdf (341.51 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01181192 , version 1 (16-10-2015)

Identifiants

  • HAL Id : hal-01181192 , version 1

Citer

Raheel Qader, Gwénolé Lecorvé, Damien Lolive, Pascale Sébillot. Probabilistic Speaker Pronunciation Adaptation for Spontaneous Speech Synthesis Using Linguistic Features. International Conference on Statistical Language and Speech Processing (SLSP), Nov 2015, Budapest, Hungary. pp.229-241. ⟨hal-01181192⟩
596 Consultations
228 Téléchargements

Partager

Gmail Facebook X LinkedIn More