ISCA Archive ICSLP 1994
ISCA Archive ICSLP 1994

The effect of the following vowel on the frequency normalization in the perception of voiceless stop consonants

Sotaro Sekimoto

The purpose of the present study is to investigate the effect of the succeeding vowel characteristics on the frequency normalization for expansion or compression of the frequency axis in the perception of voiceless stop consonants. For this purpose, two hearing experiments were carried out. Synthetic voiceless stop monosyllables that were composed of an initial noise portion and the succeeding vowel portion were used as stimuli. The noise portion was simulated by single-pole noise and the resonant frequency was systematically varied. Identification test was made to determine the categorical boundary between /t/ and /k/, or /k/ and /p/ on the continuum of the resonant frequency of the noise pole. These categorical boundaries were compared for various frequency expansion or compression ratios to assess the effect of the following vowel characteristics on the perceptual frequency normalization. The influences of the vowel identification (Exp. 1) as well as the acoustic characteristics of the following vowel (Exp. 2) were examined. The results showed that the perceptual normalization for the expansion or compression of the frequency axis depended on the formant frequencies of the following vowel. The identification of the following vowel, on the other hand, showed lesser effect on the perceptual normalization.


Cite as: Sekimoto, S. (1994) The effect of the following vowel on the frequency normalization in the perception of voiceless stop consonants. Proc. 3rd International Conference on Spoken Language Processing (ICSLP 1994), 387-390

@inproceedings{sekimoto94_icslp,
  author={Sotaro Sekimoto},
  title={{The effect of the following vowel on the frequency normalization in the perception of voiceless stop consonants}},
  year=1994,
  booktitle={Proc. 3rd International Conference on Spoken Language Processing (ICSLP 1994)},
  pages={387--390}
}