Research on Lhasa Tibetan prosodic model of journalese based on respiratory signal

In accordance with the actual development for Tibetan speech synthesis, the paper has taken news text as training corpora, analyzed the speech and prosodic features of Tibetan Lhasa dialect and confirmed the respiratory signal parameters with prosodic features. It has confirmed 6 classes of 39 dimensions context feature parameters in terms of previous prosodic structure analysis results. It uses RBF neural network to establish prosodic model and output 10 dimensions prosodic control parameters, and testing to know the predictable nature of the established model.