DocumentCode
312520
Title
A study on pitch detection using the local peak and valley for Korean speech recognition
Author
Kim, Hyung Lae ; Kim, Dae Mo ; Ryu, Young Sik ; Kim, Yung Kwon
Author_Institution
Dept. of Electron. Eng., Kon Kuk Univ., Seoul, South Korea
Volume
1
fYear
1996
fDate
26-29 Nov 1996
Firstpage
107
Abstract
The authors studied pitch detection using the local peaks and valleys for Korean speech recognition. To reduce data for detecting peaks and valleys, a center-clipping technique was adopted. The modified equation to detect peaks and valleys reduced double pitch-error and two times decimation reduced half-pitch error. The first decimation was detection of the local peaks and valleys. The second decimation was to get rid of small peaks and valleys which were not real peaks and valleys. For the experiments, the detection algorithm was developed and applied to the Korean number system
Keywords
speech processing; speech recognition; time-domain analysis; Korean; Korean number system; center-clipping technique; detection algorithm; double pitch-error reduction; half-pitch error reduction; local peaks; pitch detection; speech recognition; time domain method; valleys; Acoustic noise; Detection algorithms; Dynamic range; Equations; Frequency domain analysis; Frequency synthesizers; Signal resolution; Speech enhancement; Speech recognition; Speech synthesis;
fLanguage
English
Publisher
ieee
Conference_Titel
TENCON '96. Proceedings., 1996 IEEE TENCON. Digital Signal Processing Applications
Conference_Location
Perth, WA
Print_ISBN
0-7803-3679-8
Type
conf
DOI
10.1109/TENCON.1996.608723
Filename
608723
Link To Document