Title :
Hybrid neural predictor/stochastic trajectory models for Chinese speech recognition
Author :
Xinye, Huang ; Xiaohui, Ma ; Yuqing, Fu ; Jiren, Lu
Author_Institution :
Dept. of Radio Eng., Southeast Univ., Nanjing, China
fDate :
5/13/1999 12:00:00 AM
Abstract :
Based on stochastic trajectory models (STMs) and neural networks, a new approach to Chinese speech recognition: the hybrid neural predictor/stochastic trajectory model (STM/NP), is proposed. STM/NP makes full use of the advantages of STMs and the characteristics of the NN. The experimental results demonstrate the efficiency of the STM/NP approach
Keywords :
neural nets; prediction theory; speech recognition; stochastic processes; Chinese speech recognition; STM/NP; efficiency; hybrid neural predictor/stochastic trajectory models; neural networks;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19990587
