Pitch extraction by using autocorrelation function on the log spectrum

Extraction of fundamental frequency of the voice wave (pitch extraction) is needed in many areas of speech processing, but no established state-of-the-art algorithm exists at present. In this article, we propose anew algorithm of pitch extraction using the autocorrelation function of the log spectrum of the voice wave (ACLOS). The main point is to extract the fundamental frequency of the spectrum by calculation of the autocorrelation function of the log spectrum of the voice waves. The proposed method can improve the "gross pitch error'" (a measure of pitch extraction error) compared to all other methods used. It can improve the gross pitch error by more than 3.4% to 7.2% compared with the simple autocorrelation method in the case of an SNR less than -5 dB and has been proved to be robust by experiments, (C) 1999 Scripta Technica.