A Study on the Pitch Extraction Detection by Linear Approximation of Sub-band
The exact pitch(fundamental frequency) extraction is important in speech signal processing like speech recognition, speech analysis and synthesis. However the exact pitch extraction from speech signal is very difficult due to the effect of formant and transitional amplitude. So in this paper, the pitch is detected after the elimination of formant ingredients by flattening the spectrum in frequency region. The effect of the transition and change of phoneme is low in frequency region. In this paper we proposed the new flattening method of log spectrum and the performance was compared with LPC method and Cepstrum method. The results show the proposed method is better than conventional method.
KeywordsSpeech Signal Pitch Contour Frequency Domain Method Pitch Period Spectrum Flatten
Unable to display preview. Download preview PDF.
- Rabiner, L.R., Schafer, R.W.: Digital Processing of Speech signals. Prentice-Hall, Englewood Cliffs (1978)Google Scholar
- Paparnichalis, P.E.: Practical Speech Processing. Prentice-Hall, Inc., Englewood Cliffs (1987)Google Scholar
- Bae, M., Ann, S.: Fundamental Frequency Estimation of Noise Corrupted Speech Signals Using the Spectrum Comparison. J., Acoust. Soc. Korea 8(3) (June 1989)Google Scholar
- Lee, M., Park, C., Bae, M., Ann, S.: The High Speed Pitch Extraction of Speech Signals Using the Area Comparison Method. KIEE, Korea 22(2), 13–17 (1985)Google Scholar
- Bae, M., Rheem, J., Ann, S.: A Study on Energy Using G-peak from the Speech Production Model. KIEE, Korea 24(3), 381–386 (1987)Google Scholar
- Bae, M., Chung, I., Ann, S.: The Extraction of Nasal Sound Using G-peak in Continued Speech. KIEE, Korea 24(2), 274–279 (1987)Google Scholar