Haruko Kawasaki

Prosodic phonology and phonetics

Our colleague Charles Fillmore invented the following parable (personal communication) to characterise the two dominant ways linguists attempt to solve problems of language structure and behaviour.

Acoustic basis for universal constraints on sound sequences

A hypothesis is presented which predicts universal constraints on phoneme sequences (phonotactics) by reference to certain acoustic properties of these sequences. Two factors are hypothesized to influence these constraints. First, the magnitude of the trajectory of these soundsequences in a multidimensional acoustic space should be directly proportional to their perceptibility, and hence viability within a language. Second, the similarity of two or more such trajectories should determine the likelihood of their confusion and subsequent merger. (Unaccounted for in this analysis is unidirectional confusion and absence of soundsequences due to articulatory constraints.) The magnitude of, and distance between trajectories were computed for various sequences of initial consonant clusters and vowels. Results correctly predicted that such sequences as [bw‐], [‐wu], and [‐yi] are not as favored as [dw‐], [‐wa], [‐ya], and [‐yu]. Further, soundsequences such as [by‐] and [gw‐] were predicted to be confusable with [d‐] and lb‐l, respectively, which is in accord with known sound change processes.

Acoustic‐phonetic analysis of Japanese

How difficult would it be to convert a synthesis by rule program for English (DECtalk) to speak Japanese? What is the ideal, hopefully minimal, corpus of recorded materials that would have to be spectrally analyzed in order to derive sufficient information for a first order approximation to Japanese? We have recorded three male speakers reading a list of a little over 100 nonsense syllables, which exhaust the CV(C) syllable inventory of Japanese, and provide information about segmental contrasts. One of the speakers then read a 500‐word text to provide information on sentence‐level phenomena. We have measured segmental durations, formant motions, and deletion phenomena in broadband spectrograms of this recording, and are beginning to formulate rules for synthesis on the basis of computer analyses of the data. The approach now appears far less unreasonable than it did when we began. Some results of the analysis will be presented. [Work supported in part by an NIH grant.]

Fundamental frequency perturbation caused by voiced and voiceless stops in Japanese

The study investigates the effect of voicing of stops on the fundamental frequency (F0) of the following vowels in Japanese, particularly, how this effect interacts with phonemic accentual patterns. A speaker of Standard Japanese recorded/Can/syllables where C was/p,t,k,b,d,g,m,n/. F0 was measured for each glottal period after the consonantal release, or after the vowel onset in the case of aspirated stops. The results showed a dip in F0 around 40 ms after the release of voiced stops, while F0 generally stayed higher after voiceless stops. To examine the interaction between consonantally‐induced F0 variations and accentual patterns, the F0 contours in C1VC2V words with /t,d,n/ as C1 with either high‐low or low‐high accent were analyzed. In the high‐low accent, the low initial F0 after a voiced stop caused F0 to peak later in time than after a voiceless stop. This pattern was not observed in the low‐high accent, where the initial dip was followed by a fast increase in F0 after a voiced stop, while the pitc...

Acoustic correlates of stress in four demarcative‐stress languages

The acoustic correlates of stress in languages with “demarcative” stress (located at a fixed position relative to a word boundary) were investigated: two initial‐stress languages (Finnish, Bengali) and two final (French, Hebrew). For each language, two to three speakers produced a frame sentence, then repeated it, replacing the three‐syllable target word with /mamama/, /mimimi/, or /momomo/. Thirty reiterant versions from each speaker were analyzed for syllable duration, average intensity, intensity peak, intensity‐duration integral, and location of f0 rise and f0 peak. It was hypothesized that initial‐stress languages take advantage of higher f0 and intensity at utterance beginnings, while final‐stress languages exploit the longer duration of utterance‐final syllables. Results confirmed the association of longer duration with final stress but not of f0 and intensity with initial stress. Additionally, different syntactic/prosodic structures can change f0 patterns. For example, in French, one frame sentence caused f0 to rise in the last syllable, while another did not. This raises questions about the relative importance of different acoustic parameters for (a) the categorization of demarcative‐stress languages, and (b) lexical versus phrasal prominence.The acoustic correlates of stress in languages with “demarcative” stress (located at a fixed position relative to a word boundary) were investigated: two initial‐stress languages (Finnish, Bengali) and two final (French, Hebrew). For each language, two to three speakers produced a frame sentence, then repeated it, replacing the three‐syllable target word with /mamama/, /mimimi/, or /momomo/. Thirty reiterant versions from each speaker were analyzed for syllable duration, average intensity, intensity peak, intensity‐duration integral, and location of f0 rise and f0 peak. It was hypothesized that initial‐stress languages take advantage of higher f0 and intensity at utterance beginnings, while final‐stress languages exploit the longer duration of utterance‐final syllables. Results confirmed the association of longer duration with final stress but not of f0 and intensity with initial stress. Additionally, different syntactic/prosodic structures can change f0 patterns. For example, in French, one frame sentenc...

Acoustic basis for universal constraints on phoneme combinations

Phonological studies reveal many cross‐language similarities in permissible/impermissible phoneme sequences. We attempt to account for such universally attested phonotactic constraints by reference to two aspects of acoustic properties of these sound sequences. First, the magnitude of acoustic modulation in these sound sequences should be directly proportional to their perceptual saliency, which in turn would affect their viability in languages. Second, the similarity between two or more sound sequences should determine the likelihood of confusion and their susceptibility to merger. The magnitude of, and distance between formant trajectories were computed for various sequences of stops, liquids, glides, and vowels. The results correctly predicted the infrequent occurrence of [bw‐], [‐wu], [‐yi] as opposed to frequently observed [‐wa], [‐ya], [‐yu]. Such sequences as [gw‐] and [dw‐] are predicted to be confusable with [b‐], and [by‐] is likewise confusable with [d‐]. The results also indicated that the com...

An acoustic analysis of connected Japanese digits

This study investigated the acoustic characteristics of coarticulatory phenomena in connected Japanese digits. Its aim was to discover acoustic phonetic regularities that could be used in a knowledge‐based continuous speech recognition system. The data consisted of 840 six‐digit strings spoken by seven males and three female native speakers of Japanese. A variety of intra‐ and inter‐digit coarticulatory phenonema were examined using synchronized displays of digitized waveforms, wideband spectrograms, total energy, low‐frequency energy, and zero crossings rate. Among the questions assessed were: whether allophonic processes often cited in the phonological literature are readily observable in the acoustic signal, whether they are consistently observed across tokens and speakers, and whether seemingly free variants occur truly randomly or are conditioned by such factors as phrasal structure, accentual pattern, and speaking rate. The allophonic processes examined included: devoicing/shortening of high vowels,...

An acoustic study of Japanese coronal fricative clusters resulting from vowel devoicing

Earlier studies [S. Shattuck‐Hufnagel, V. Zue, and J. Bernstein, J. Acoust. Soc. Am. Suppl. 1 64, S92 (1978), V. Zue and S. Shattuck‐Hufnagel, Proc. 9th ICPS 2, 215 (1979)] examining English fricative clusters [s∫] and [∫s] across word boundaries (e.g., in “gas shortage” and “tunafish sandwich”) showed that the direction of assimilatory palatalization is predominantly anticipatory: [s∫] often becomes [∫∫], but [∫s] does not. This study investigates the cross‐linguistic validity of such a tendency. Four speakers of standard Japanese casually spoke sentences containing fricative clusters [∫s] and [s∫] resulting from devoicing/deletion of an intervening high vowel [i] or [u]. Preliminary observations reveal complex patterns not totally consistent with those for English: anticipatory palatalization in [s∫] in Japanese does not seem as robust a process as in English. Furthermore, perseveratory palatalization, i.e., ∫s→∫∫, and both anticipatory and perseveratory depalatalization, i.e., ∫s→ss, s∫→ss, all of which are relatively rare in English, are found in Japanese data. The results suggest a diffculty in explaining the previous English data in terms of peripheral articulatory control.

Articulation of [i]'s

The articulation of the palatal vowel [i] in various phonetic environments was studied in one speaker of English and one speaker of Japanese using the technique of dynamic contact palatography. The utterances examined were of the form C1V1C2V2, where V1 or V2 was [i][u] or [a] and the other vowel always [i]. C2 was any of a variety of consonants proper to the languages involved. Preliminary results suggest that the articulation of [i] is influenced in complex ways by both consonantal and vocalic environment. For example, “front” consonants such as [n] serve to preserve the canonical contact pattern for [i] whereas “back” consonants such as [k] permit greater deviation from the canonical form due to coarticulation with adjacent vowels. Other things being equal, [i] seems to dominate in the coarticulation between vowels in the speech of the Japanese speaker. A momentary reduction in the extent of palatal contact during the consonant in an [ipi] utterance was noted. This is apparently due to passive enlargem...