论文信息 - Phoneme Dedicated ANN Improves Segmental Duration Model

Phoneme Dedicated ANN Improves Segmental Duration Model

The Phoneme Dedicated Artificial Neural Network (PDANN) segmental duration model consists of a set of ANNs trained specifically for each phoneme segment in order to avoid miscellaneous influence of different types of phoneme segments. Therefore, each ANN is dedicated to predict the duration of a specific phoneme segment. Objective and subjective measurements of the performance of the PDANN model were compared with those of a typical ANN model using the same input features and database. The results indicate a slight, but clear, perceptually perceived preference towards the PDANN.

João Paulo Teixeira | Diamantino Freitas

[1] João Paulo Teixeira. A prosody model to TTS systems , 2004 .

[2] Diamantino Freitas,et al. Phonetic events from the labeling the european portuguese database for speech synthesis, FEUP/IPBDB , 2001, INTERSPEECH.

[3] Jan P. H. van Santen,et al. Assignment of segmental duration in text-to-speech synthesis , 1994, Comput. Speech Lang..

[4] Plínio A. Barbosa,et al. A model of segment (and pause) duration generation for Brazilian Portuguese text-to-speech synthesis , 1997, EUROSPEECH.

[5] W. Nick Campbell. Predicting segmental durations for accommodation within a syllable-level timing framework , 1993, EUROSPEECH.

[6] Plínio Almeida Barbosa. Caractérisation et génération automatique de la structuration rythmique du français , 1994 .

[7] D. Klatt. Linguistic uses of segmental duration in English: acoustic and perceptual evidence. , 1976, The Journal of the Acoustical Society of America.