On the Complex Regulating the Voiced-Voiceless Distinction I

An investigation is described of the perceptual and acoustical differences between voiceless and voiced consonants in Dutch. The experiments reported included synthesized, normal and whispered speech. The results are compared with those found in the literature for other languages. Parallels between the results of the various investigations could be established, although quantitative differences apply to different languages. It is shown that a number of perceptual and acoustical cues are present, such as: sound intensity and duration of fricative noise, duration of the consonant and duration of the preceding vowel, voicing and the moment of voice onset, intonation pattern and amplitude as well as duration and frequency range of the formant transitions of the adjoining vowels. To establish a connection between these parameters an articulatory model of the difference voiced-voiceless is proposed. Essential features in the construction of this model are the observed acoustical symptoms and physiological data as reported in the literature. The model is based on the assumption that the difference voiced-voiceless is due to the presence or absence of activity of the pharyngeal constrictor muscle. This muscle influences the volume of the pharynx and the position of the larynx, and consequently the pressure drop across the glottis and the vibration condition of the vocal cords.