An Experimental Analysis of the Relative Importance of Pitch, Quantity, and Intensity as Cues to Phonemic Distinctions in Southern Swedish

Abstract

This paper, following introductory comments on previous experiments involving prosodic features of speech, reports an experiment designed to indicate the relative importance of pitch, quantity, and intensity as cues for syllabic prominence in Southern Swedish. Spectrographic analysis of the particular minimal pair used in this experiment revealed that one, or any combination, of three possible prosodic features, pitch, quantity, and intensity, could account for the contrast in meaning of the pair, the written representation of which is HÄLSA PA. Stimuli were prepared in which all possible binarily opposed combinations of pitch, quantity, and intensity in the last syllable were represented. Additional stimuli were prepared to test whether the contrastive pitch level of the first syllable might also contribute to the identification of the minimal pair. Native speakers of Swedish were asked to identify each stimulus item as either greet or visit, the two concepts which the written form HÄLSA PA represented. Analysis of the results showed that pitch provided the primary cue for identifications. Significantly, the pitch of the first syllable was the most potent cue, overriding opposing cues of pitch, quantity, and intensity on the final syllable. Accordingly, the conclusion was drawn that perception followed primarily the first syllable, that subjects made their identifications on the basis of the pitch of the first syllable, and that they could not be dissuaded in their identifications by contradictory information present in the prosody of the final syllable. It could not be established with certainty which of the two remaining cues, quantity or intensity, was the more important.

Acoustic Correlates of Stress in Nepali Language

“…The results of the study concluded that there is significant difference between fundamental frequency and intensity and no significant difference in duration of words. The primary acoustic correlates of stress in various language such as Polish (Jassem et al, 1968), English (Bolinger,1958) and French (Rigault, 1962) is different in fundamental frequency where as for Swedish (Westin, Buddenhagen & Obrecht 1966), Italian (Bertinetto, 1980), and Estonian (Lehiste,1968) had duration as the major correlates. Recent studies also show that, Kannada (Savithri, 1999), Konkani (Kumar & Bhat 2009), Tamil (Balasubramanian, 1981), had duration as the major correlates for acoustic stress, whereas according to Sitapati (1936) and Srinivas (1992), intensity was reliable acoustic correlate of stress in Telugu language.…”

“…Studies done earlier has shown that, the acoustic correlates of stress vary from language to language and stressed syllables are usually associated with one or more of the following properties i.e., raised fundamental frequency, increased loudness, greater duration, and different vowel qualities (Liberman, 1960). In languages such as English (Bolinger, 1958), Polish (Jassem, Morton, Steffen, 1968) and French (Rigault, 1968), fundamental frequency was observed as the primary acoustic correlate of emphatic stress, whereas duration was reported to be major correlate in languages such as Swedish (Westin, 1966), Italian (Bertinetto, 1980) and Estonian (Lehiste, 1968). In the Indian context, research on acoustic correlates of stress in words were carried out in Kannada (Savithri, 1999), Konkani (Kumar, 2009), Tamil (Balasubramanian, 1981) which indicated increased duration in stressed conditions.…”

‘Native’ and ‘non‐native’ perception of stress in Singapore English

“…Testing his hypotheses on each parameter in turn in a series of experiments on British English revealed that fundamental frequency ranks as the most dominant perceptual cue in British English, followed by duration, intensity, and finally vowel quality. Various other studies (Bolinger ; Morton & Jassem ; Westin et al ; Awedyk ; Eek ; Beckman ; Lass ) have also shown that different perceptual cues are employed for different languages or varieties of the same language. The study of stress in SgE therefore cannot be conducted by simply adopting or assuming the cues of stress established for another variety of English.…”