Vocal tract resonances tracking by auxiliary vector particle filters  

作　　者：WANG Yebin ZHAO Heming 

机构地区：[1]School of Electronics and Information Engineering, Soochow University Suzhou 215006

出　　处：《Chinese Journal of Acoustics》2011年第1期105-114,共10页声学学报（英文版）

基　　金：supported by the National Natural Science Foundations of China(60572076)

摘　　要：An auxiliary vector particle filter was proposed to present the vocal tract resonances （VTRs） tracking. It uses particle filter based on a version of state-space model that describes the characteristics of speech signal. The speech model consists of a target-guided dynamic function and a non-linear prediction mapping from resonance frequencies and bandwidths to LPC cepstra （LPCC）. There are two characteristics in the proposed method. First, particle filtering technique is put forth to solve the non-linear problem of speech model. Second, an auxiliary vector, embedded in the state function of speech model, is applied to incorporate the most current observations and to generate the proposal distribution of particle filter. The experimental results show that this method is able to track the VTRs of continuous speech utterance efficiently with a small number of particles and able to solve the problem of spurious peaks and merging peaks.An auxiliary vector particle filter was proposed to present the vocal tract resonances （VTRs） tracking. It uses particle filter based on a version of state-space model that describes the characteristics of speech signal. The speech model consists of a target-guided dynamic function and a non-linear prediction mapping from resonance frequencies and bandwidths to LPC cepstra （LPCC）. There are two characteristics in the proposed method. First, particle filtering technique is put forth to solve the non-linear problem of speech model. Second, an auxiliary vector, embedded in the state function of speech model, is applied to incorporate the most current observations and to generate the proposal distribution of particle filter. The experimental results show that this method is able to track the VTRs of continuous speech utterance efficiently with a small number of particles and able to solve the problem of spurious peaks and merging peaks.

分 类 号：TN713[电子电信—电路与系统] TP277[自动化与计算机技术—检测技术与自动化装置]