Volume 13, Issue 2 (2022)
LRR 2022, 13(2): 105-136 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
hosseinpoor damirchian R, Nourbakhsh M. The Effect of Voicing on Constriction Duration, Voice Duration and Vowel Duration in Stop and Fricative Consonants of Turkish Language in Tabrizi Dialect. LRR 2022; 13 (2) :105-136
URL: http://lrr.modares.ac.ir/article-14-50113-en.html
URL: http://lrr.modares.ac.ir/article-14-50113-en.html
1- M.A students of Linguistics, Linguistics Department, Fculty of Literature, Alzahra University, Tehran, Iran , r.hosseinpoor@student.alzahra.ac.ir
2- Associate Professor, Linguistics Department, Fculty of Literature, Alzahra University, Tehran, Iran
2- Associate Professor, Linguistics Department, Fculty of Literature, Alzahra University, Tehran, Iran
Abstract: (2569 Views)
Voicing is one of the most important phonological features in distinguishing consonants in many languages. In this study, we investigated the temporal parameters such as constriction duration, voice duration, and vowel duration in stop and fricative consonants of Azeri Turkish language in intervocalic position (VCV). The main purpose of this study is to study stop and fricative consonants in Tabrizi dialect. An attempt was made to answer the question of how the voicing of stop and fricative consonants of Turkish language affects the temporal parameters. Twelve selected words of Turkish common words were repeated by fourteen Tabrizi speakers. They were produced three times in citation form. The results showed that temporal parameters are considered as potential cues in distinguishing voiced and voiceless Turkish consonants. The results related to stop consonants also showed that only voice duration and vowel duration were effective in the voicing contrast. The results for the closure duration confirmed that there is no significant difference between voiced and voiceless stops and this cue can not be used as a distinguishing cue to voicing in Azeri Turkish stops.
1. Introduction
Glottis and its condition is one aspect of describing consonants; So the vibration in vocal cords during consonant production produces voiced sounds unless the produced sound is voiceless. Voicing is one of the phonological features that has been discussed in most languages and has certain acoustic cues. Some of these cues are defined as temporal features, which include constriction duration, voice duration, and the preceding vowel duration. In other words, these temporal features are considered cues for distinguishing obstruent consonants. Steriade (1997, pp. 6-7) offers a list of cues to voicing and their distribution and believes that changing stop consonants’ position in a syllable can cause a change in the number of voicing cues. closure duration and closure voicing are two essential cues to study stop consonants voicing. Moreover, there are sixteen acoustic features classified into three groups including pre-closure, closure, and post-closure features (Lisker, 1986). Lisker (1986) argues that these features are potentially perceptual cues to voicing distinction in the intervocalic position. This means that closure duration and voice duration are associated with closure, and vowel duration relates to pre-closure features.
Turkish language is one of the most common languages in the Altaic language family which has the largest number of speakers (Crystal, 1987, p. 307). The southwestern branch is one of the main branches of Turkish language, which includes Azerbaijani, Turkish, and Turkmen (Hayat, 2001, p. 8). In general, the Turkish language of Azerbaijan can be divided into two dialects: the northern dialect in the Republic of Azerbaijan and the southern dialect in Iranian Azerbaijan (Johanson, 1998). Among the common non-Persian languages in Iran, Azerbaijani Turkish with 15 to 20 million native speakers has the highest number of speakers (Crystal, 2010) and also among 26 types of Azeri dialects, Baku, Guba, Lankaran, Shirvan, and Tabrizi dialects are common in East Azarbaijan region (Heyat, 2001). This language is agglutinative and has 24 consonants and 9 vowels. Vowels are /ɑ, ɯ, o, u, æ, e, i , œ, y/. Also out of 24 consonants, there are 6 stop consonants / b, p, d, t, ɡ, ɟ / and 9 fricative consonants /f, v, s, z, ʃ, ʒ, x, ɣ, h/ in this language. Additionally, there are two other stops /c/ and /k/ that are used only in the Southern dialect and in loan words (Ghaffarvand Mokri and Warner, 1396).
The main purpose of this study is to examine stop and fricative consonants of Azeri Turkish in the Tabrizi dialect, acoustically. So that these consonants will be evaluated by measuring acoustic cues such as closure duration, voice duration, and vowel duration in intervocalic position. This is an experimental study that will be performed by designing a production test. In this study, we aim at providing an answer to the following question: whether voicing has a significant effect on the duration of acoustic cues in stop and fricative consonants in Azeri Turkish language? We assume that the duration of above-mentioned cues is significantly influenced by voicing feature of obstruents.
2. Literature Review
There are several studies done on many languages evaluating acoustic cues in obstruent consonants such as stop closure duration (Shen et al., 1987; Lisker, 1957; Nikrahi, 2012; Razavi Najafabadi & Nourbakhsh, 2013), fricative constriction duration (Jongman, 2000; Nartey, 1982; Klatt, 1976; Baum &Blumstein, 1987; Rahimi, 2013), affricate closure and frication duration (Hosseinpoor Damirchan & Nourbahsh, 2021), voice onset time (Jahan, 2009; Ünal-Lugacev et al, 2018) and vowel duration ( Ladefoged, 2006, p. 58; Chen, 1970; Raphael, 1972; Warren & Marselen-Wilson, 1989).
3. Methodology
Fourteen Azeri native speakers participated in the production experiment. Twelve Turkish words containing 4 stops /d, t, b, p/ and 8 fricatives /z, s, ʒ, ʃ, ɣ, x, v, f/ were produced in the intervocalic position between vowels /ɑ/. Participants were in a quiet room and produced syllables three times in citation form and also they were asked to repeat with a short interval between each word. Thus, 504 data were obtained (14 speakers × 12 words × 3 repetitions). Praat voice Analysis Software Version 6 .1. 30 were used for acoustic analysis of data. All measurements were done manually considering both waveforms and spectrograms. Also, SPSS software Version 23 was employed for statistical analysis.
4. Results
4-1. Stop closure duration and Frication duration
Closure duration in stop consonants provides two main acoustic information associated with the voicing feature. Examining this cue showed that the length of closure is not significantly longer in viceless stops than voiced (p ≥0.05). The mean closure duration is 93.87 and 98.86 for voiced and voiceless stops respectively. Therefore, closure duration can not be considered a proper acoustic cue to contrast voicing in Azeri Turkish stops.
The results of frication length measurement showed that voiceless fricatives (mean duration: 161.7) are longer than voiced fricatives (mean duration: 91.58). Mean and standard deviation are shown in table 2 for each fricative. The longer duration of voiceless fricatives is significant (p ≤ 0.05) and leads to the result that length of frication is an important cue to fricatives voicing. Table 1 shows descriptive statistics of stops closure duration and fricative constriction duration.
Table 1.
Descriptive statistics of constriction duration in stops and fricatives
4-2. Stop and fricative voice duration
Mean and standard deviation are shown in table 2 for stop and fricative consonants. The average voice duration obtained for voiced stops is 57.79 which is longer than voice length in voiceless stops with 25.13, therefore the significant difference between voiced and voiceless stops (p ≤ 0.05) demonstrates that voice length is the main cue to voicing recognition.
The overall average voice duration for voiced fricatives is longer than the voiceless with 86.33 and 29.93 respectively. In other words, voiceless fricatives have shorter voice length and this difference is considerably significant (p ≤ 0.05). Thus, voice duration is considered an important cue to voicing category of fricatives.
Table 2.
Descriptive statistics of voice duration in stops and fricatives
4-3. Vowel duration preceding stops and fricatives
Mean and standard deviation of vowel length before stops and fricatives are shown in table 3 and indicate that length of vowel before voiced consonants are longer than voiceless ones. The significant difference of voiced and voiceless stops and fricatives (p ≤ 0.05) shows that the duration of vowel precesing obstruens is an essential cue to the voicing distinction.
Table 3
Descriptive statistics of vowel duration before stops and fricatives
6. Conclusion
In this study, three acoustic cues in stop and fricative consonants voicing were investigated for speech production in Azeri Turkish language spoken in Tabriz. Except for stop closure duration, the other cues showed significant differences between voiced and voiceless obstruents. Hence, frication duration, voice duration, and vowel duration before obstruents are potential cues to voicing distinction in Turkish language in Tabrizi dialect.
1. Introduction
Glottis and its condition is one aspect of describing consonants; So the vibration in vocal cords during consonant production produces voiced sounds unless the produced sound is voiceless. Voicing is one of the phonological features that has been discussed in most languages and has certain acoustic cues. Some of these cues are defined as temporal features, which include constriction duration, voice duration, and the preceding vowel duration. In other words, these temporal features are considered cues for distinguishing obstruent consonants. Steriade (1997, pp. 6-7) offers a list of cues to voicing and their distribution and believes that changing stop consonants’ position in a syllable can cause a change in the number of voicing cues. closure duration and closure voicing are two essential cues to study stop consonants voicing. Moreover, there are sixteen acoustic features classified into three groups including pre-closure, closure, and post-closure features (Lisker, 1986). Lisker (1986) argues that these features are potentially perceptual cues to voicing distinction in the intervocalic position. This means that closure duration and voice duration are associated with closure, and vowel duration relates to pre-closure features.
Turkish language is one of the most common languages in the Altaic language family which has the largest number of speakers (Crystal, 1987, p. 307). The southwestern branch is one of the main branches of Turkish language, which includes Azerbaijani, Turkish, and Turkmen (Hayat, 2001, p. 8). In general, the Turkish language of Azerbaijan can be divided into two dialects: the northern dialect in the Republic of Azerbaijan and the southern dialect in Iranian Azerbaijan (Johanson, 1998). Among the common non-Persian languages in Iran, Azerbaijani Turkish with 15 to 20 million native speakers has the highest number of speakers (Crystal, 2010) and also among 26 types of Azeri dialects, Baku, Guba, Lankaran, Shirvan, and Tabrizi dialects are common in East Azarbaijan region (Heyat, 2001). This language is agglutinative and has 24 consonants and 9 vowels. Vowels are /ɑ, ɯ, o, u, æ, e, i , œ, y/. Also out of 24 consonants, there are 6 stop consonants / b, p, d, t, ɡ, ɟ / and 9 fricative consonants /f, v, s, z, ʃ, ʒ, x, ɣ, h/ in this language. Additionally, there are two other stops /c/ and /k/ that are used only in the Southern dialect and in loan words (Ghaffarvand Mokri and Warner, 1396).
The main purpose of this study is to examine stop and fricative consonants of Azeri Turkish in the Tabrizi dialect, acoustically. So that these consonants will be evaluated by measuring acoustic cues such as closure duration, voice duration, and vowel duration in intervocalic position. This is an experimental study that will be performed by designing a production test. In this study, we aim at providing an answer to the following question: whether voicing has a significant effect on the duration of acoustic cues in stop and fricative consonants in Azeri Turkish language? We assume that the duration of above-mentioned cues is significantly influenced by voicing feature of obstruents.
2. Literature Review
There are several studies done on many languages evaluating acoustic cues in obstruent consonants such as stop closure duration (Shen et al., 1987; Lisker, 1957; Nikrahi, 2012; Razavi Najafabadi & Nourbakhsh, 2013), fricative constriction duration (Jongman, 2000; Nartey, 1982; Klatt, 1976; Baum &Blumstein, 1987; Rahimi, 2013), affricate closure and frication duration (Hosseinpoor Damirchan & Nourbahsh, 2021), voice onset time (Jahan, 2009; Ünal-Lugacev et al, 2018) and vowel duration ( Ladefoged, 2006, p. 58; Chen, 1970; Raphael, 1972; Warren & Marselen-Wilson, 1989).
3. Methodology
Fourteen Azeri native speakers participated in the production experiment. Twelve Turkish words containing 4 stops /d, t, b, p/ and 8 fricatives /z, s, ʒ, ʃ, ɣ, x, v, f/ were produced in the intervocalic position between vowels /ɑ/. Participants were in a quiet room and produced syllables three times in citation form and also they were asked to repeat with a short interval between each word. Thus, 504 data were obtained (14 speakers × 12 words × 3 repetitions). Praat voice Analysis Software Version 6 .1. 30 were used for acoustic analysis of data. All measurements were done manually considering both waveforms and spectrograms. Also, SPSS software Version 23 was employed for statistical analysis.
4. Results
4-1. Stop closure duration and Frication duration
Closure duration in stop consonants provides two main acoustic information associated with the voicing feature. Examining this cue showed that the length of closure is not significantly longer in viceless stops than voiced (p ≥0.05). The mean closure duration is 93.87 and 98.86 for voiced and voiceless stops respectively. Therefore, closure duration can not be considered a proper acoustic cue to contrast voicing in Azeri Turkish stops.
The results of frication length measurement showed that voiceless fricatives (mean duration: 161.7) are longer than voiced fricatives (mean duration: 91.58). Mean and standard deviation are shown in table 2 for each fricative. The longer duration of voiceless fricatives is significant (p ≤ 0.05) and leads to the result that length of frication is an important cue to fricatives voicing. Table 1 shows descriptive statistics of stops closure duration and fricative constriction duration.
Table 1.
Descriptive statistics of constriction duration in stops and fricatives
| Mean | SD | Min | Max | ||
| Stop | voiced | 93.87 | 16.88 | 58 | 145 |
| voiceless | 98.86 | 23.26 | 48 | 148 | |
| fricatives | voiced | 91.58 | 24.38 | 50 | 166 |
| voiceless | 161.7 | 25.54 | 104 | 218 | |
Mean and standard deviation are shown in table 2 for stop and fricative consonants. The average voice duration obtained for voiced stops is 57.79 which is longer than voice length in voiceless stops with 25.13, therefore the significant difference between voiced and voiceless stops (p ≤ 0.05) demonstrates that voice length is the main cue to voicing recognition.
The overall average voice duration for voiced fricatives is longer than the voiceless with 86.33 and 29.93 respectively. In other words, voiceless fricatives have shorter voice length and this difference is considerably significant (p ≤ 0.05). Thus, voice duration is considered an important cue to voicing category of fricatives.
Table 2.
Descriptive statistics of voice duration in stops and fricatives
| Mean | SD | Min | Max | ||
| Stop | voiced | 57.79 | 27.37 | 10 | 112 |
| voiceless | 25.13 | 11.45 | 7 | 63 | |
| fricatives | voiced | 86.33 | 26.32 | 17 | 166 |
| voiceless | 29.93 | 10.89 | 10 | 80 | |
4-3. Vowel duration preceding stops and fricatives
Mean and standard deviation of vowel length before stops and fricatives are shown in table 3 and indicate that length of vowel before voiced consonants are longer than voiceless ones. The significant difference of voiced and voiceless stops and fricatives (p ≤ 0.05) shows that the duration of vowel precesing obstruens is an essential cue to the voicing distinction.
Table 3
Descriptive statistics of vowel duration before stops and fricatives
| Mean | SD | Min | Max | ||
| Stop | voiced | 114.87 | 23.43 | 60 | 184 |
| voiceless | 96.94 | 23.02 | 59 | 170 | |
| fricatives | voiced | 135.74 | 32.34 | 76 | 234 |
| voiceless | 101.33 | 23.11 | 53 | 159 | |
6. Conclusion
In this study, three acoustic cues in stop and fricative consonants voicing were investigated for speech production in Azeri Turkish language spoken in Tabriz. Except for stop closure duration, the other cues showed significant differences between voiced and voiceless obstruents. Hence, frication duration, voice duration, and vowel duration before obstruents are potential cues to voicing distinction in Turkish language in Tabrizi dialect.
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |