MARYTTS HMM-based voices for the Blizzard Challenge 2012
Marcela Charfuelàn · 2012
This paper describes the first participation of MARY TTS HMM-based voices in a Blizzard challenge. An architecture for synthesis of expressive speech based on the MARY TTS system and sentiment analysis of text is proposed. The creation of several HMM-based voices in different styles using audiobook data is described. Preliminary results on perception of different voice styles and the appropriateness of a given style for a given sentence are presented. The latest developments in the open source MARY TTS 5.0 are briefly described. Index Terms: speech synthesis, parametric speech synthesis, expressive speech, sentiment analysis, signal processing.