This paper describes the evaluation of lifters with respect to their frequency spectra. A lifter is the quefrency window for the cepstrum and is used for eliminating the pitch components of the speech signal to obtain the spectral envelope. We have already suggested that comb lifters are useful for separating the close formants even if the pitch frequency is high [G. Oyama et al., IEEE‐ICASSP 78, 19–22 (April 1978)]. It is desirable that the comb lifters should have a sharp main lobe and small side lobes in the frequency spectrum. To make the main lobe sharp, it is necessary that the quefrency of the first stopband of the comb lifter should be high, though the quefrency depends upon the pitch frequency. The lifter should have gentle slope along the quefrency axis to reduce the side lobes in the frequency spectrum. Therefore the envelope of the comb lifter should decrease gradually with an increase in the quefrency. To suppress the peaks due to the pitch, the stop bands in the comb lifter should have certain width, because the pitch frequency of natural speech usually fluctuates. Frequency spectra of various types of comb lifters are investigated on the basis of these considerations.
Skip Nav Destination
,
,
Article navigation
November 1978
August 11 2005
Evaluation of lifters by frequency response Free
Gen Ooyama;
Gen Ooyama
Research Center for Applied Information Sciences, Tohoku University, Sendai, 980 Japan
Search for other works by this author on:
Shigeru Katagiri;
Shigeru Katagiri
Research Center for Applied Information Sciences, Tohoku University, Sendai, 980 Japan
Search for other works by this author on:
Ken'iti Kido
Ken'iti Kido
Research Center for Applied Information Sciences, Tohoku University, Sendai, 980 Japan
Search for other works by this author on:
Gen Ooyama
Shigeru Katagiri
Ken'iti Kido
Research Center for Applied Information Sciences, Tohoku University, Sendai, 980 Japan
J. Acoust. Soc. Am. 64, S160 (1978)
Citation
Gen Ooyama, Shigeru Katagiri, Ken'iti Kido; Evaluation of lifters by frequency response. J. Acoust. Soc. Am. 1 November 1978; 64 (S1): S160. https://doi.org/10.1121/1.2003950
Download citation file:
113
Views
Citing articles via
Focality of sound source placement by higher (ninth) order ambisonics and perceptual effects of spectral reproduction errors
Nima Zargarnezhad, Bruno Mesquita, et al.
A survey of sound source localization with deep learning methods
Pierre-Amaury Grumiaux, Srđan Kitić, et al.
Variation in global and intonational pitch settings among black and white speakers of Southern American English
Aini Li, Ruaridh Purse, et al.
Related Content
Displaying speech as vocal tract area function pictures
J. Acoust. Soc. Am. (August 2005)
A low bit rate vocoder based on an improved cepstral method
J. Acoust. Soc. Am. (August 2005)
Conversion of the vocal tract shape for spectral warping by a PARCOR analysis—synthesis system
J. Acoust. Soc. Am. (August 2005)
Some comparisons between the articulation rates of LPC and diphone or monophone‐based synthesis by rules
J. Acoust. Soc. Am. (August 2005)
LPC speech at 1200 bits per second using optimized frame repeat
J. Acoust. Soc. Am. (August 2005)