Fundamental frequency extraction method based on the p‐th power of amplitude spectrum with band limitation
Tetsuya Shimamura, Hiroshi Takagi · Electronics and Communications in Japan (Part III Fundamental Electronic Science) · 2007
Abstract The extraction of the fundamental frequency of speech is an important issue in many speech processing fields. In this paper, we propose a fundamental frequency extraction method based on the power of the amplitude spectrum in the bandwidth where the fundamental frequency is present. This method varies the power of the amplitude spectrum depending on the signal‐to‐noise ratio (SNR) of the input speech signal in each analysis frame. Therefore, in a white noise environment, this method is applicable to a wide range of speech signals from a high signal‐to‐noise ratio to a low signal‐to‐noise ratio. By conducting preliminary tests using true vowels, we derive the equation representing the relationship between the power of the amplitude spectrum and the signal‐to‐noise ratio. This is also applied to real continuous speech, and the superiority of the proposed method is clearly demonstrated in comparative tests with conventional methods. © 2007 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 90(5): 19–30, 2007; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/ecjc.20069