VOICE SOUNDS, sound made by the human voice, which— as used in English and other European speech—can be produced in two different forms, namely (a) unvoiced, i.e., breathed or whispered speech, and (b) voiced speech.
In unvoiced speech the vocal chords are more or less separated so that the air from the lungs passes continuously between them. In voiced speech the vocal chords are brought closely together so that the forcible passage of the air between them sets up a rhythmical vibration of the chords which causes the air to enter the vocal cavities in correspondingly rhythmical puffs.
Vowel Sounds.—Until recently the accepted theory was that sounds like i as in eat, e as in men, ae as in hat, were each due to two separate resonances, that a as in calm, was produced in some voices by a single resonance, and in others by two resonances, and that 3 as in all, ou as in no, u as in who, were all due to single resonance (I). According to Helmholtz's theory (2) the lower series of double resonances is set up in the cavity behind the tongue, and the upper in a tubular neck formed by the tongue and lips, while in the case of the single resonant sounds, the whole oral cavity is supposed to act as a single resonator. Graham Bell, the inventor of the telephone, and K. J. Lloyd, both held that all vowel sounds were due to double resonance (3).
Recent observations (4) have confirmed the views of Graham Bell and Lloyd, and shown that all the English vowel sounds depend primarily on two characteristic resonances—the one formed in the cavity in front of the tongue—i.e., between the point
of nearest approach of the tongue to the palate and the lips— and the other in the cavity behind the tongue—i.e., between the same point and the vocal chords. Both cavities behave as Helm holtz resonators connected in series. Each can be independently tuned, by varying its capacity or by varying the size of its orifice to air and (or) to the adjoining resonator. Increase of capacity lowers the resonant pitch ; of orifice raises it.
The following table shows typical resonances observed by ear in the whispered vowel sounds of the writer's voice, using the Southern English "Public School" pronunciation.
Vowel Resonances The resonances are given in musical notation and in number of complete vibrations per second (c' = middle c on the piano forte, about 256 vibrations per second). In the series from 9 to v inclusive, an additional high frequency component between 2,169 and 2,732 appears. The upper resonance of i to ov can be varied over a range of 5 to 8 semitones, and u and v by 10 semi tones, while the lower resonances can all be similarly varied over a range of about 8 semitones (the other resonance, in each case, being kept nearly constant) without losing the vowel character. Comparable resonances have been found—using purely instru ment methods—by Crandall and Sacia (5).