Towards automatic phonetic segmentation for TTS

Towards automatic phonetic segmentation for TTS

Phonetic segmentation is an important step in the development of a concatenative TTS voice. This paper introduces a segmentation process consisting of two phases. First, forced alignment is performed using an HMM-GMM model. The resulting segmentation is then locally refined using an SVM based bounda...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Rendel, A., Sorin, A., Hoory, R., Breen, A.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Accuracy
Acoustics
Adaptation models
Data models
Hidden Markov models
Phoneme alignment
Phonetic segmentation
Text to speech
Training
Vectors
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Phonetic segmentation is an important step in the development of a concatenative TTS voice. This paper introduces a segmentation process consisting of two phases. First, forced alignment is performed using an HMM-GMM model. The resulting segmentation is then locally refined using an SVM based boundary model. Both the models are derived from multi-speaker data using a speaker adaptive training procedure. Evaluation results are obtained on the TIMIT corpus and on a proprietary single-speaker TTS corpus.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2012.6288926