Published online by Cambridge University Press: 22 April 2026
[Recent experiments with synthetic speech have shown that the perception of voiceless stops depends in large measure on the context, not on the characteristics of the consonants alone. The same phenomenon was investigated in actual speech by recording voiceless stops on tape before certain vowels, cutting them away, and splicing them back before other vowels. The change in context caused certain instances of [k] to be perceived as [p] or [t]. The experiment shows that in actual speech the context of an initial voiceless stop is an important factor in its perception; the results agree strikingly with those obtained for synthetic speech. It is further suggested that aspiration ought properly to be regarded as part of the following vowel rather than of the preceding consonant.]
1 A shorter version of this paper was presented before the Modern Language Association of America in Boston, December 1952. The research on which the paper is based was supported in part by the Signal Corps, the Air Materiel Command, and the Office of Naval Research, and in part by the Carnegie Corporation.
2 Liberman, Delattre, and Cooper, The rôle of selected stimulus-variables in the perception of the unvoiced stop consonants, American journal of psychology 1952.497-516.
3 This experiment was carried out by Liberman, Delattre, and Cooper; see the reference in fn. 2. The work with actual speech was a direct outgrowth of their experiment with synthetic speech; it was undertaken to test the validity of inferences about actual speech based on the results obtained with synthetic speech.