Title :
A time warping neural network
Author_Institution :
Dept. of Electr. Eng., Stanford Univ., CA, USA
Abstract :
A method is proposed to improve any temporal pattern recognition system by time warping each pattern before presentation to the recognition system. The time warping function for a pattern is generated by repeated local application of a neural network to sections of the pattern. The output of this neural network is the slope of the warping function, and the internal weight parameters are trained by a gradient descent learning rule which attempts to minimize the recognition system´s error. Experimental results show that this method can improve recognition of vowel phonemes
Keywords :
learning (artificial intelligence); neural nets; speech recognition; time warp simulation; gradient descent learning rule; internal weight parameters; speech recognition; temporal pattern recognition system; time warping neural network; training; vowel phonemes; warping function; Cognition; Dynamic programming; Equations; Feedforward neural networks; Iron; Neural networks; Optimization methods; Pattern recognition; Sampling methods; Speech recognition;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on
Conference_Location :
Detroit, MI
Print_ISBN :
0-7803-2431-5
DOI :
10.1109/ICASSP.1995.479700
