Title :
A new method for the voiced/unvoiced decision based on pattern classification theory
Author :
Zhijie, Zhou ; Guangrui, Hu
Author_Institution :
Shanghai Jiaotong Univ., China
Abstract :
The voicing decision for a speech segment is a very important subject in speech processing. It is the key issue to improve the quality of speech processing. In this paper, a voicing decision algorithm using pattern classification is discussed. Some rough achievements from computer simulation are presented. The correct rate for the voicing decision of our method reaches 97.4 percent
Keywords :
decision theory; pattern classification; speech recognition; pattern classification theory; quality; speech processing; speech segment; unvoiced decision; voicing decision algorithm; Bit rate; Computer simulation; Human voice; Linear discriminant analysis; Loudspeakers; Pattern classification; Speech coding; Speech processing; Speech recognition; Vectors;
Conference_Titel :
Signal Processing, 1996., 3rd International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-2912-0
DOI :
10.1109/ICSIGP.1996.567362
