diva模型中定时和预测功能的分析word格式论文.docxVIP

AbstractIn recent years, it is the key idea of designing the speech synthesis system to describe and simulate the functions corresponding to the associated brain areas and cerebellum areas, which are involved in speech production and acquisition at the level of neuroanatomy and neuropsycology. Around this topic, great efforts were made in researching and discussing the computational model of speech production and acquitision. The most outstanding model is the DIVA (Directions Into Velocities of Articulators) model which is proposed by the professor Frank Guenther and his colleges in speech lab, Boston University. The DIVA model is a neural network model describing the process of speech production and acquisition to product word, syllable, and phonemes by controlling a simulated vocal tract.First, the developing process of the DIVA model is introduced. The associated theory of DIVA model is illustrated in detail, which include the basic concept and construction of the DIVA model, and the learning phase and imitation phase.Second, the region of interest, ROI is introduced, and the speech related ROI in brain, other ROI in brain and the cerebellum ROI was in detail respectively.Third, in order to solve the cerebellum timing function in DIVA model, after analyzing the feedforward control subsystem, the RSL model is introduced. Then the DIVA model is modified by using RSL cerebellum system, so as to adjust the cerebellum delay in the subsystem. The simulation results demonstrates that the modification makes the cerebellum timing function more accurate, and also reduces the error in speech production.At last, on the basis of introducing the hireachical state feedback control model, the DIVA model is compared to the HSFC model. And the HSFC model is applied to the DIVA model. The experiment result shows that the modification can effectively improve the accuaracy of prediction further to decrease the error in speech production.This article aims to improve the precision and