An Rbfn-Based system for speaker-independent speech recognition

Faculty Engineering Year: 1995
Type of Publication: Theses Pages: 169
Authors:
BibID 10956530
Keywords : System identification    
Abstract:
A speaker-independent isolated-word small vocabulary system is developed for applications such as voice-driven menu systems.The design of a cascade of recognition layers is present ed .Several feature sete are compared. phone recognition is performed using a radial basis function network (RBFN)dynamic time warping (DTW)is used for word recogntion.The TIMIT database is used to design and test the automatic speech recognition (ASR)system. 
   
     
PDF  
       
Tweet