Title :
Estimation of probability density function and evaluation by vowel recognition
Author :
Nakagawa, Seiichi ; Ono, Yoshiyuki ; Hur, Kangin
Author_Institution :
Dept. of Inf. & Comput. Sci., Toyohashi Univ. of Technol., Japan
Abstract :
We propose an estimation method of probability density function using a feedforward neural network and apply it to the vowel recognition task. The network, trained with target vectors made from ratios among numbers of training data in a partial input, space, can represent a probability density function of the data.
Keywords :
estimation theory; feedforward neural nets; learning (artificial intelligence); probability; speech recognition; statistical analysis; vector quantisation; feedforward neural network; probability density function estimation; speech recognition; target vectors; training data; vector quantisation; vowel recognition; Computer science; Density functional theory; Feedforward neural networks; Feedforward systems; Neural networks; Probability density function; Probability distribution; Smoothing methods; Speech recognition; Training data;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.714168
