how do foreign students apprehend indonesian speech sounds?...

Running Head: HOW DO FOREIGN STUDENTS APPREHEND INDONESIAN SPEECH

InternationalConferenceonLanguage,Literature,CultureandEducation17th&18thOCTOBER,2015

68

3 ICLLCE 2015 20 Ichwan Suyudi

How Do Foreign Students apprehend Indonesian Speech Sounds? (An Experimental Phonetic Study)

Ichwan Suyudi*, Debyo Saptono

Gunadarma University - Indonesia *Corresponding e-mail : [email protected]

ABSTRACT

Prosody refers to intonation of the spoken language in linguistic field. It is a broad concept, consists of some features such as tempo, time, accent, stress, timbre (Arnfield, 1994). These research goals are to (1) know the characteristic of phonetically system such as vocal pronunciation of foreign speakers; (2) describe the foreign speaker pronunciation based on the produced articulation pronunciation and articulation place. The research method is experimental phonetic (IPO approach) using PRAAT software. The result research shows various sounds in some analyzed samples. The foreign learners of Indonesian language often make errors in producing Indonesian words. Those errors are caused by some difficulties, such as (1) difficulty in learning how to produce foreign sounds in speech organ itself; (2) the sound distribution of the proper sound spoken for a word and in the context of the spoken word; (3) the fluency, that is a ability in pronunciation of all set of sound (group of sounds) easily and fast; (4) the relation between conventional pronunciation and spelling.

Keyword: prosody, IPO, PRAAT, foreign student

Backgrounds

The prosody of mother language (LI) often influences the pronunciation of the second language (L2). This influence of LI is called as language transfer. The negative transfer often happens when a person has already had accent of certain language, pronunciation of other language, such as foreign speakers of Indonesian language which often speaks Indonesian language with the accent of their mother languages. This transfer can be in the segmental and super segmental level (prosody) because there are various changes between their mother languages and Indonesian Language.

Sound or super segmental element can be differentiating into some elements, they are the stress about the hard and low of the sound. The distinctive element can be seen from the significant differences between English which distinctively stress, but in Indonesia language, it is not. For example, the word, ‘blackboard’ can be stressed in the black element so that the meaning is ‘the board to write’; if the stress is in the board element, so that the meaning will be ‘the board with black color’. In Indonesian language, the word ‘old people’ can be stressed in the people or old, the meaning will remain the same. The sound with most resonance space is the vocal sound. Thus, the syllabic sound or peak sound is the vocal sound. As an example, the word [and]. This word consists of [d], [a], and [n] sounds. And the [d] and [n] sound are the consonant sounds, while the [a] sound is the vocal one. The [a] sound in that word is the peak syllabi and loudness, because as mentioned above, the [a] sound is as vocal produced to have bigger resonance space.

HOW DO FOREIGN STUDENTS APPREHEND INDONESIAN SPEECH


69

Goerge(1977) states two anticipation steps in solving the language error, they are giving the time for correction, directing the attitude towards the target language, identifying the undesired form, selecting the undesired form, learning every error, determining the further strategy in improving learning. Other researcher, Lightbown and Nina Spada (1999) decreases the error level in foreign speaker that is to get the proper information about the phonetic at that time.

The foreign speakers learned Indonesian Language is expected have the ability to speak as one of the aspect of fluency in productive language. The spoken language is more function in daily activity. Speaking is language form using articulation or words used to deliver the intention (Hurlock 1981). Dardjowidjojo (1995) frequently examines the error in Indonesian language learning and explains that the error of foreign speakers because of the influence of mother language, minimum acquisition of Indonesian language principles and limitation of Indonesian vocabulary. The frequent errors are the different of utterance tools, the place of articulation and the articulator which is able to make different meaning significantly. This difference does not only refer to meaning of phonemic but more to the difference in phonetic so that it may be new findings in phonetic ability of foreign speaker articulator in Indonesian language learning. This research goal is to know the prosody (super segmental element) transferred into pronunciation of Indonesian language words by the foreign speakers. As detail, these research goals are:

1. To know the phonetic system characteristics of words as the vocal pronunciation by the foreign speakers.

2. To describe the foreign speaker pronunciation based on the produced articulation pronunciation and the place of articulation.

3. To describe the tendency of foreign speaker utterance change in Indonesian language.

Research Methods

The language sound data in the form of wave is analyzed using Praat program used to analysis the phonetic data. PRAAT, used in some previous researches (Veraci, 2011, Pranowo, 1984, Silvana, 2000, et al) is open source software found by Paul Boersma and David Weenink from Phonetic Science Department University of Amsterdam used to analysis the sound reconstruction analysis flexible.

The analysis procedure can be done in ranks. In detail, the sound data of every sound is processed by using Praat program to every data in label level.

The research objects :

1. The sound data or language sound from the informants with recording technique. 2. The foreign speakers learning Indonesian language (BIPA) 3. The high intensity of Indonesian language use 4. The age of 20-40 years old 5. Having complete articulator

The respondents are selected by purposive sampling from the BIPA students. The data processing is done by (1) sound extraction, (2) segmentation, (3) doing the frequency measurement and extracting the result measurement in the data terminal. The frequency measurement is done by adopting the IPO theory (Instituute voor Perceptie onderzoek) (Collier, Cohen, 1990), (4) copying data in Praat Picture or close copy.



70

The Results and Discussion This research is focused in the experimental analysis of sound data by using Praat software. The result analysis from the informants is as follow:

• The ‘datang’ <Coming>Analysis Horizontal –F2 value, Vertical –F1 Value Coming

From the three informants, the pattern close to native is b74 although there is difference in position and height of tongue in beginning and final sounds. The 54a informant pattern is very different, the height and position of tongue just move a little bit, the height of tongue is between 600-800 Hz, while the range of the height of tongue is from 250 until 1000.

• The ‘gitar’ <Guitar> Analysis Horizontal –F2 value, Vertical –F1Value Guitar

Based on the figure in the left, the a43 informant pattern is the most close to the Native similarity. While, the de3 informant pattern is the less close to the native similarity. The sound of /i/ causes the different other informant. It seems that when the bb1 informant gives /i/ sound, the tongue is closer to tongue palate while the native tongue is less close to the palate. The final sound of /r/ is also by different sound of de3 informant. The /r/ sound is done by moving the tongue getting put but more to the palate. This is different with the native one, the tongue is less closer to the palate.

54a,1613.94

3,660.782

54a,1436.42

7,667.502

54a,1761.12,710.777

54a,1401.66

7,723.239

54a,1390.42

1,671.444 b74,

2203.074,

398.24

b74,1931.00

6,815.915

b74,1914.78

4,849.079

b74,1410.52

9,1004.30

5b74,

1446.828,

554.179126,

1724.124,

369.719

126,1672.75

5,689.562

126,1698.33

5,739.647

126,1609.23

3,994.789126,1722.72

2,708.007

95c-native,1716.18

4,536.937

95c-native,1761.20

5,786.301

95c-native,1697.50

4,992.556

95c-native,1274.92

6,364.392

95c-native,1648.75

9,473.467

54ab7412695c-native

bb1,1987.841,

467.199

bb1,2029.57,

762.072

bb1,1811.51,

510.137

bb1,1515.484,

678.795

bb1,1515.484,

678.794

a43,1984.419,

574.775

a43,1796.667,

364.141

a43,1900.202,

923.645

a43,1419.076,

972.833

a43,1736.66,

850.704de3,2352.839,

311.643

de3,2349.759,

357.23

de3,2174.604,

585.728

de3,1496.794,

886.839

de3,1664.852,

858.517

d2a-native,1803.356,

846.804d2a-

native,1785.217,

298.475


967.536


1070.849


680.725

bb1

a43

de3

d2a-native



71

• The ‘Insyaf’ <Conscious >Analysis Horizontal –F2 Value, Vertical –F1 Value

Based on the figure in the left, it seems that 40d shows the pattern that is almost the same with the Native about the height of tongue and the back and forth movement of the tongue. The 71b informant show the difference, mainly in the first and final sound of /i/ and /f/. the ce2 informant is the most different informant among others, mainly in producing sound of /sy/, the tongue is getting out and down compared to the native.

• The ‘pagar’ <Hedge> Analysis Horizontal –F2 value, Vertical –F1 value

Based on the figure in the left, there is no pattern from the three informants which is the same with Native. The native positions the tongue getting out and the position of tongue closer to the mouth palate. The d77 informant seems to be different in producing the sound of /r/ by positioning the tongue getting out from the mouth and the height of tongue is getting far from the mouth palate while the native’s tongue is positioned getting out from the mouth and the position of tongue is getting close to the palate.

Results and Discussion

Laksman (1995) explains about the identification of physical sound characteristic in Indonesian language. Laksman explains the formant frequency is the frequency of complex sound waved periodically and strengthened in sound cavity such as pharyngeal, mouth and vocal cord. This formant frequency gives certain patterns about the vocal sound and some consonants, especially the frequency of the two first formants (F1 and F2) (pp 93).

ce2,2192.41

9,397.633ce2,

2094.43,337.3

ce2,2193.59

3,1190.59

ce2,2018.42

5,510.617

ce2,1846.51

4,655.947

ce2,1849.27

1,638.57971b,1758.36

1,383.892

71b,1551.34

2,378.158

71b,2295.61

6,918.03471b,

2069.24,642.909

71b,1662.63

1,951.19

71b,1418.88

2,873.798

40d,2011.83

1,292.269

40d,2173.82

5,311.271

40d,1732.63

4,1300.45

1

40d,1845.24

1,593.209

40d,1499.43

8,783.748

40d,1961.82

8,991.118

0db-Native,1754.29

8,513.14

0db-Native,1803.07

3,399.43

0db-Native,1987.11

5,929.9970db-Native,2196.13

1,442.941

0db-Native,1605.97

9,852.266

0db-Native,1817.87

3,1214.55

7ce2

71b

40d

0db-Native

4aa,1480.504,674.741

4aa,1602.367,653.315

4aa,1701.17,854.5

4aa,1464.628,860.848

4aa,1627.257,828.187

756,1431.946,958.684

756,1597.518,885.999

756,1686.218,487.774

756,1768.605,754.539756,

1804.004,710.061

d77,2168.239,784.213

d77,1592.6,725.519

d77,1774.853,481.818

d77,1625.155,722.572

d77,1586.019,856.858

bf2-native,1302.81,509.95

bf2-native,1458.812,1074.117

bf2-native,1727.872,405.575

bf2-native,1680.695,326.75

bf2-native,1735.509,428.083

4aa756d77bf2-native



72

Some researchers are done to get the value of formants in Indonesian language by this following explanation. The physical

characteristic of Indonesian language (Lapoliwa)

The physical characteristic of Indonesian language in batak Toba tribe (van Zanten)

The physical characteristic of Indonesian language in Javanese tribe (van Zanten)

F1 F2 F3 F1 F2 F3 F1 F2 F3 [i] 250 2450 3300 291 2190 3075 319 2174 2912 [u] 300 1750 2500 377 829 2449 378 856 2515 [a] 590 1590 2830 764 1365 2362 750 1345 2480 [ə] 410 2080 2920 389 1794 2623 519 1367 2501 [o] 496 869 2569 544 871 2533 The source of this research data is the word by foreign and Indonesian speakers. The vocal spoken can be positioned in the first, medium and final. The example can be seen from the following table. Word First position Medium position Final position ahad [a] [had] alfa [al] [fa] The result research can be seen in this figure below. The above values, then, are got the median and the data is below. First position Medium position Final position The physical

characteristic of Indonesian language (Lapoliwa)

F1 F2 F1 F2 F1 F2 F1 F2 [i] 396 1929 710 2095 367 2002 250 2450 [u] 548 1403 565 1498 300 1750 [a] 890 1491 770 1541 788 1572 590 1590 [ə] 617 1420 614 1679 410 2080 [o]

Based on the comparison of this result research with Lapoliwa research, the results

are as follow. 1. The F1 [i] value in first and final position is closer to the F1 Lapoliwa value, while

the F2 value in the medium position is close to the F2 Lapoliwa value. 2. The F1 [u] value is higher than the F1 Lapoliwa value. this means that when

producing the [u] sound, the position of tongue is getting farer from the mouth palate compared to the Lapoliwa result research. While, the F2 [u] value is lower than the F2 Lapoliwa value. this means that the tongue moves to go inner compared to the Lapoliwa result research.

3. The F1 [a] value is higher than the F1 Lapoliwa value. this means that when producing [a] sound, the tongue position is getting farer from the mouth palate than the Lapoliwa result research. While, the F2 [a] value in the medium and final position is getting closer to the F2 Lapoliwa value because of the less far differences.

4. The F1 [ə] value is higher than F1 Lapoliwa value. this means that when producing the [ə] sound, the tongue position is getting farer to the mouth palate



73

compared to the Lipoliwa result research. This means that the tongue moves to get inner than the Lipoliwa result research.

5. Overall, only F2 [a] values in medium and final position are the closest to the F2 Lapoliwa value.

Conclusion Basically, the sound system in Indonesian language is the same with the sound

system in English. Although there are some vocals in English which are not in Indonesian language. in English, stress is very important because it influences the meaning of a word, while in Indonesian language, it doesn’t influence the meaning of word. Besides, in English, there is aspirated sound, meaning that it is the sound with blowing air followed when it is spoken.

The Indonesian language learners often make some errors in pronunciation of Indonesian language words. These errors are caused by some difficulties, are:

1. The difficulty in learning how to make foreign sounds with utterance organ itself. 2. The distribution of sound, that is which sound is proper when spoken in such

word or sentence, and in what context the sound is spoken. 3. The fluency, that is the ability to speak all set of sounds (group of sounds) easily

and fast. 4. The relation between conventional pronunciation and spelling. The above errors

are the most common error faced by the learners. In learning foreign language, especially Indonesian language, the speaker really has to understand that the language sound system is different with English. There are some sounds in English that are not in Indonesian language, such as vocal, consonants, diphthong, or other form of sounds. Besides, in Indonesian language, stress, length, and intonation don’t give influence to the meaning of word. These problems can be solved by trying to know how the way to speak or pronounce a word and practice to pronounce and also practice to listen how the native speakers pronounce directly or indirectly.

References Boersma, P. Weenink, D. 2004. PRAAT: doing phonetics by computer, v.4.2.2.06. http://www.fon.

Hum.uva. nm/praat/ Frederick Williams. 1973. Normal Aspect of Speech, Hearing, and Language. New Jersey: Prentice

Hall. Gerhard, D., 2003. Pitch Extraction and Fundamental Frequency: History and Current Techniques,

Department of Computer Science University of Regina, Regina Hayward, Katrina. 2000. Experimental Phonetics. Harlow: Longman J’t Hart, J., R. Collier, and A. Cohen. 1990. A Perceptual Study of Intonation: An Experimental-

phonetic Approach to Speech Melody. Cambrige: Cambrige University Press. Ladefoged, Peter. Elements of Acoustic Phonetics. Chicago: University of Chicago Press. Laksman, Myrna. The Perception of the Place of Stress In Indonesian Words and Phrases. 1995, Tap

San Khoa Hoc, Annals of Hochiminh City University Lapoliwa, Hans. 1988. Pengantar Fonologi I: Fonetik. Jakarta: Departemen Pendidikan dan

Kebudayaan. Lehiste, Ilse. 1970. Suprasegmentals. Cambridge: The MIT Press.Moain, Amat Juhari. 1989. Lieberman, Philips. 1972. Speech Acoustics and Perception. Indianapolis. The Bobb-Merrill

Company. Nooteboom, Sieb. 1999. “The Prosody of Speech: Melody and Rhythm”. Dalam Hardcastle, William J. and John Laver. 1999

how do foreign students apprehend indonesian speech sounds?...

Documents