Regional Variation in Phonetic Responses to Time Pressure in Tehrani and Isfahani Varieties

Abstract

AbstractIn languages and varieties, the exact locations of the alignment of f0 transition points differ from those of the segmental string (Ladd, 2006). The differences of inter-languages or inter-varieties are sometimes so small that it is not possible to identify a deep representation. We conducted a production experiment with 40 speakers of two regional varieties of Persian language. The aim of the present study was to investigate the strategies used by the speakers of the varieties in response to the time pressure caused by reduced sonorant rime of target words by adopting autosegmental-metrical theory (Pierrehumbert, 1980). For this purpose, 20 Isfahan speakers and 20 Tehrani speakers were asked to repeat the three target words, which were accompanied by a decrease sonorant rime, in the positions of nuclear pitch accent twice. Three prosodic parameters, such as contour conditionm alignment, and duration of the segments containing pitch accents, were mainly considered. The results showed that in each contour condition, the duration of the sonorant rime of the nuclear accented word was varied. In the phonetic domain, the range of adjustments fell into three broad categories. First, the speakers of Isfahani variety took more time by increasing the duration of the rime in the fall-rise condition while retracting the peak of the fall-rise contour. However, the speakers of Tehrani variety worked harder by compressing contours in the fall contour. Both speakers of the two varieties took shortcuts by truncating responses of the final pitch movements. The responses were shown to be dialect-specific and contour-specific.Keywords: Duration of Sonorant Rime, Pitch Accent, Timing or Alignment, Time Pressure IntroductionSpeakers of different varieties produce the same intonation contours, which are phonetically differently. One of the factors influencing on the realization of contours is the time available for execution. Movement f0 requires the minimum amount of time to be executed. Productions of the targets can be compromised under tonal crowding (Pierrehumbert, 2000). The magnitude of time pressure in intonation-only languages like Dutch and English firstly depends on the available spaces between tonal targets, for example between a prenuclear and nuclear pitch accents or between a nuclear pitch accent and a phrasal boundary. There is less time available for executing the prenuclear and nuclear accents in a skittish kitten, which has shorter vowels and less voicing as compared to a roaring lion, with an identical syllable structure but more voicing. The available time in the black cat is, in turn, less compared to the situation of a skittish kitten, which follows a lack of intervening unaccented syllables (accent clash) (Silverman and Pierrehumbert, 1990) and a shorter space between the nuclear accent and the intonational phrase edge. The shorter the stretch of sonorant segmental material is, the more the realization of nuclear and boundary tones will be under pressure (Grabe, 1998b). Also, tonal crowding will be more problematic in fast speech compared to normal or slow speech because the time to produce tonal targets is reduced due to shorter segmental durations (Caspers and van Heuven, 1993 for Dutch).Hanssen (2017) mentions that speakers use 3 economizing mechanisms for F0 adjustments in response to time pressure. The first is “truncation”, whereby speakers produce an incomplete version of the contour, leaving the speed of the f0 movement intact. The second is “compression”, whereby the full contour is produced but at a higher speed. The third one is that speakers can respond to time pressure by increasing the duration of the available sonorant material so as to create more time for the execution of the pitch movement. Speakers may add duration when a movement is physiologically more difficult to produce. Figure 1 shows the economic strategies that speakers use in response to time pressure: Schematic illustrations of truncation, undershoot, and compression (columns) for falls, rises, and fall-rises (rows). The gray uninterrupted lines represent an f0 contour on a long sonorant portion, while the black interrupted lines represent an f0 contour on a short sonorant portion. The arrows and scissors illustrate how the f0 contours have been affected by time pressure. An undershoot can affect a fall-rise in two ways; it affects either the dip between two high tones or the initial high tone.This paper reported what strategies speakers adopt in response to time pressure due to limited sonorant material in IP-mid nuclear contour. In this study, we sought to respond to the strategies used by the speakers of two species (Isfahani and Tehrani) in response to the time pressure caused by reduced limited sonorant material and find the answer to this question: what strategies do the speakers of the two varieties of languages ​use to deal with time pressure? Material and MethodsTo conduct this research, we presented 3 utterances that had different sonorant rimes to 20 Isfahani (10 males and 10 females) and 20 Tehrani language speakers (10 males and 10 females) to produce them during a production experiment. This research investigated the regional variation in the realization of intonation contours in Isfahani and Tehrani varieties in the framework of autosegmental-metrical theory. The autosegmental-metrical theory is basically based on the principle that the intonation is completely systematic and has a phonological structure; that is, the intonation forms a hierarchical prosodic structure in the form of a sequence of tonal units (L, H, and a combination of these two tones). The intonation contour (a type of contour that shows the pitch excursions of an utterance over the time) is represented linearly through autosegmental tones, each of which is aligned with a specific syllable or a specific place in a segmental group. The combination of these tones determines the prosodic structure of the speech.Each dataset contains several contour pitches that are phonologically identical, but differ in rise, fall-rise, and rise type. A total of 240 sentence samples (1 sentence * 3 target words * 40 speakers * 2 repetitions) were recorded. Then, using Praat speech processing software (Boersma and Winink , 2019), the intonational pattern of the extracted sentences and the intonational variables were coded manually. To observe and study the changes of F0 along the pitch contour, the average values were​ obtained from the difference between the minimum base frequency and the maximum base frequency of the sonorant rime. In this study, for the nuclear pith accent state, the 3 variables of pitch contour, duration, and alignment were measured. The statistical analysis of the data was carried out by using SPSS. Discussion of Results & Conclusion The purpose of this study was to investigate the effects of time pressure on the parameters of pitch contour, alignment, and duration. By examining the research data, th following result was obtained: In the two Isfahani and Tehrani varieties, the three target words had a significant effect on the sonorant rime (p<0.05). The sonorant rimes of coda in the fall and rise contours, were more in "Yɑm" compared to "Yɑr" and "Yɑb" in both varieties, respectively. However, in the fall-rise contour, the sonorant rime of coda was higher in "Yɑr" compared to "Yɑm" and "Yɑb", respectively. The statistical results related to the time pressure between the two contours, simple contour (fall), and complex contour (fall-rise) showed that the type of contour in terms of simplicity and complexity had an effect on the average sonorant rime of coda so that in Isfahani and Tehrani varieties, the average sonorant rimes of coda in the fall-rise contour were longer than those of the fall contour. Also, adaptation of alignment under time pressure due to complexity of the contour showed that the speakers started the fall-rise contour earlier than the fall contour. The data related to the investigation of the effect of the boundary of the next intonational phrase on the time pressure in thr fall, fall-rise, and rise contours in the two varieties showed that the type of sonorant rime of the word had a significant effect on the peak alignment in the fall contour. This issue confirmed that the sonorant rime in the fall-rise contour was significantly longer than that of the fall contour. In addition, the influence of the contour type on the relative alignment in Tehrani and Isfahani varieties revealed that the peak in the fall-rise contour started earlier than the peak in the fall contour. In both varieties, the peak in the relative alignment was influenced by the word, while the speakers adjusted the peak according to the sonorant rime and placed about 20-30% of it at the end of coda, which indicated late occurrence of the H tone. Comparison of the peak delays between the words in the fall-rise contour in Isfahani and Tehrani varieties showed that the differences in peak delays in Yɑr-Yɑm, Yɑr-Yɑb, and Yɑm-Yɑb were significant. Yɑr-Yɑm and Yɑr-Yɑb were not significant in the word Yɑm-Yɑb. Also, the results showed that the type of sonorant rime had a significant effect on the relative peak alignment in Isfahani and Tehrani varieties (p<0.05). The comparison of the relative peak alignments in Isfahan and Tehrani varieties in the words "Yɑr, Yɑm, and Yɑb" revealed that the difference in the relative peak of Yɑr-Yɑm was not significant, but it was significant in the words of Yɑr-Yɑb and Yɑm. The comparison of the relative valley data between the words in the rise contour showed that in Isfahani and Tehrani varieties, the words had a significant effect on the relative valley (p<0.05). Comparing the relative alignments of the valley between the words "Yɑr/Yɑm/Yɑb" in Isfahani variety showed that the difference in the relative valley of the words Yɑr-Yɑm was not significant, but this difference was significant in the words of Yɑr-Yɑb and Yɑm-Yɑb. In Tehrani variety, the comparison of the relative alignments of the words "Yɑr/Yɑm/Yɑb" showed that the difference in the relative valleys of Yɑr-Yɑm and Yɑm-Yɑb was significant, but this difference in the words Yɑr-Yɑb was not significant.The results indicated that the speakers of Isfahani variety spent more time in the fall-rise contour and used the undershoot phenomenon, leading to an earlier occurrence with lower scaling on the affected units. While the speakers of Tehrani variety worked harder by compressing the contours in the fall contour. Both speakers of the two varieties took shortcuts by truncating responses of the final pitch movements. According to the observations of the authors in the main data, it seemed that the speakers’ choices in the diversity of contours were quite systematic. According to PFitzinger (2009), the choice of diversity of contour is special and systematic.

Keywords

• duration of sonorant rime
• pitch accent
• timing or alignment
• time pressure