Title :
A new efficient pitch-tracking algorithm
Author :
Li, Bo ; Li, Ying-ying ; Wang, Cheng-you ; Tang, Chao-jing ; Zhang, Er-yang
Author_Institution :
Sch. of Electron. Sci. & Technol., Nat. Univ. of Defense Technol., Changsha, China
Abstract :
The quality of the synthetic speech generated by harmonic models for speech is essentially influenced by the quality and the faultlessness of the pitch measurement. In this paper, we present a new technique for detecting the pitch of speech. We use the pitch harmonical autocorrelation (PHA) for the initial pitch detection. Then we use the autocorrelation of the speech frequency to refine the pitch. Compared the other method, the algorithm presented in this paper is more easy and more accurate. The efficient technique allows us to precisely find out the pitches of speech.
Keywords :
harmonic analysis; speech processing; harmonic models; pitch harmonical autocorrelation; pitch tracking algorithm; speech frequency; speech pitch detection; speech pitch measurement; speech quality; Acoustic noise; Autocorrelation; Chaos; Equations; Frequency estimation; Intelligent systems; Neural networks; Signal analysis; Speech analysis; Speech processing;
Conference_Titel :
Robotics, Intelligent Systems and Signal Processing, 2003. Proceedings. 2003 IEEE International Conference on
Print_ISBN :
0-7803-7925-X
DOI :
10.1109/RISSP.2003.1285744
