Ninth International Conference on Spoken Language Processing

Pittsburgh, PA, USA
September 17-21, 2006

Multi-Domain Text-to-Speech Synthesis by Automatic Text Classification

Francesc Alías, Joan Claudi Socoró, Xavier Sevillano, Ignasi Iriondo, Xavier Gonzalvo

Ramon Llull University, Spain

This paper describes a multi-domain text-to-speech (MD-TTS) synthesis strategy for generating speech among different domains and so increasing the flexibility of high quality TTS systems. To that effect, the MD-TTS introduces a flexible TTS architecture that includes an automatic domain classification module, which allows MD-TTS systems to be implemented by different synthesis strategies and speech corpus typologies. In this work, the performance of a corpus-based MD-TTS system is subjectively validated by means of several perceptual tests.

Full Paper

Bibliographic reference.  Alías, Francesc / Socoró, Joan Claudi / Sevillano, Xavier / Iriondo, Ignasi / Gonzalvo, Xavier (2006): "Multi-domain text-to-speech synthesis by automatic text classification", In INTERSPEECH-2006, paper 1579-Tue3BuP.3.