论文信息 - Improving AMDF for pitch period detection

Improving AMDF for pitch period detection

Pitch period is the important parameters of speech recognition and speech synthesis. Pitch period detection has been focus in the field of audio processing research. Traditional AMDF-based algorithm and its improved version, LV-AMDF-based algorithm easily leads to the double error or half error, and so on in the pitch detection. To solve these problems, AMDF, LV-AMDF function characteristics and shortcomings in pitch detection are analyzed, the parameters compensation AMDF pitch detection algorithm is proposed in this article to reduce semi-frequency, double-frequency errors often appear in the pitch detection and improve the detection accuracy. Experimental results show that its pitch detection accuracy is better than AMDF and LV-AMDF.

Guo Shize | Kang Guangyu

[1] Ting,et al. Study on automatic prediction of sentential stress for Chinese Putonghua Text-to-Speech system with natural style , 2007 .

[2] Zhao Yong-zhen. Study on automatic prediction of sentential stress with natural style in Chinese , 2006 .

[3] P. Boersma. ACCURATE SHORT-TERM ANALYSIS OF THE FUNDAMENTAL FREQUENCY AND THE HARMONICS-TO-NOISE RATIO OF A SAMPLED SOUND , 1993 .

[4] Xu Li-min. Research of pitch detection based on LPC-residual cepstrum , 2007 .

[5] Tetsuya Shimamura,et al. A modified cepstrum method for pitch extraction , 1998, IEEE. APCCAS 1998. 1998 IEEE Asia-Pacific Conference on Circuits and Systems. Microelectronics and Integrating Systems. Proceedings (Cat. No.98EX242).

[6] Zhaohui Wu,et al. A pitch-based rapid speech segmentation for speaker indexing , 2005, Seventh IEEE International Symposium on Multimedia (ISM'05).