Perceptual and acoustic correlates

The nature of stress is well understood. Stressed sounds are those on which the speaker expends more muscular energy. This usually involves pushing out more air from the lungs by contracting the muscles of the rib cage and perhaps increasing the pitch by the use of the laryngeal muscles. The extra activity may result in the sound having greater length. There may also be increases in the muscular activity involved in the articulatory movements (Ladefoged 2003).

When there is an increase in the amount of air being pushed out of the lungs, there is an increase in the loudness of the sound produced. Some books define stress simply in terms of loudness, but this is not a very useful definition if loudness is considered to be simply a matter of the amount of acoustic energy involved. We know that some sounds have more acoustic nergy than others because of factors such as the degree of mouth opening. Loudness is not by itself an efficient device for signalling the location f accent in English. When they are said on a monotone and without un-ue lengthening of accented syllables, it is difficult to distinguish by loud-ess alone im'port and below from 'import and 'billow, words in which different accentual patterns are not associated with qualitative differences in he vowels.

A much more important indication of stress is the rise in pitch, which may or may not be due to laryngeal activity {Gimson 1972, Cruttenden 2001).

There is a final factor when discussing stress in English. We saw that a syllable in English is either stressed or unstressed. If it is stressed, it can be at the center of an intonational pitch change so that it receives a tonic accent, which might be said to raise it to a primary level of stress. If it is unstressed, it can have a full vowel or a reduced vowel. In some views, a reduced vowel implies that there is a lower level of stress, but in the view expressed here it is not a matter of stress but of vowel quality. We also saw that there are pairs of words, such as (an) 'insult and (to) in'sult, that differ only in stress. What happens when these words happen to lose their stress because of a heavy stress elsewhere in the sentence? Consider such a pair of sentences as He needed an increase in price and He needed to increase the price. The answer is that the stress difference is the relative lengths of the syllables. A stressed syllable is pronounced with a greater amount of energy than an unstressed syllable, and this difference maybe manifested simply in the length of the syllable.

Thus, from the listener's point of view, or in terms of perception, the prominence of a stressed syllable is achieved through greater length, greater loudness, a change in pitch (normally higher pitch but a slide would do as well) and vowel quality. Not all languages use these phonetic cues, but the English and the Russian languages are reported to possess all the possibilities of which two or three may be used simultaneously. There is evidence that they may have trading relations, i.e. one may compensate for the other.

The acoustic correlates of the prosodic features of length, loudness and pitch are duration, intensity and fundamental frequency. On the articulatory level greater loudness is achieved by subglottal muscular adjustment which will create an acoustic effect of increased amplitude. Laryngeal muscular adjustment will create increased tension in the vocal folds which will result in increased fundamental frequency, i.e. pitch. However, the auditory sensation of loudness will increase here as well. Thus there is no simple one-to-one correspondence between a certain articulatory gesture and an auditory effect. But the connection is there: experimental evidence suggests that we perceive speech sounds by refer­ence to the articulatory movements which we ourselves would have to make in order to produce the same sounds. Listeners compare their "inner" knowledge as speakers to the input from the acoustic signal in making stress judgements.

Of special importance is the time given to the effort: all the perceptual features of length, pitch and loudness may depend on the timing of the syllable in a word. In Russian there is a definite correlation in length between the stressed and the preceding unstressed syllables. Cf. the relative length of vowels in the stressed and the preceding syllables of the Russian word хоро'шо. The time ratio is reported to be equal to 1:2:3. Some schol­ars believe that the quality of vowel articulation depends on the time given, thus if there is little time for articulation the speaker "undershoots" the desired point of articulation, and the vowel is reduced both in length and quality.

There is one segmental feature here, vowel quality, which corresponds acoustically to spectrum. Other segmental features may have additional value for a stressed syllable distinction: strong aspiration of fortis plosives /p, t, k/ before a stressed vowel in English, a glottal stop before a stressed vowel in other Germanic languages.

Thus we can conclude that while accent is principally achieved by pitch change, sometimes assisted by extra loudness, among unaccented syllables some will be more prominent than others due to the quality and quantity of the vowels at the centre. Long vowels and diphthongs are generally more prominent than short vowels, while among the short vowels themselves, /i, u, 3/ (when unaccented) are the least prominent and are often referred to as reduced vowels as opposed to other full vowels (as far as prominence is con­cerned, syllabic consonants are considered to be sequences of/э/ plus /1, m, n, rj/ and hence are equivalent to reduced vowels). Indeed the reduced vowels are so lacking in prominence that they have a high frequency of occurrence in unaccented as opposed to accented syllables. But a long vowel in an unaccented syllable is sometimes longer than a vowel in an adjacent accented syllable, e.g.pillow ['pilau], ally ['aelai], phoneme [Тэишт], record ['reko:d] etc.

Дата добавления: 2014-01-11