Fourth ISCA ITRW on Speech Synthesis
August 29 - September 1, 2001
The text normalisation (tokenisation and pre-processing) and the linguistic analysis (grapheme-to-phoneme conversion and prosody generation) within the Elan text-to-speech (TTS) industrial system are described. This system comprises eight languages: French, English, Spanish, Brazilian Portuguese, German, Russian, Italian and Polish. Applications involving output normalisation are also presented.
Bibliographic reference. Boula de Mareüil, Philippe / Soulage, Benoît (2001): "Input/output normalisation and linguistic analysis for a multilingual text-to-speech synthesis system", In SSW4-2001, paper 109.