Name: Hong Li Student Number:1904874 Subject: Essay for Key issues in Linguistics (HPEN7004) Q3. Critique the application of Praat in doing phonetic analysis Acquisition of English Intonation of Chinese EFL speakers 1. Introduction Phonetics is the basic and important part of the teaching system of English learning. To Learn a language, one only needs to learn it 50% Grammar and 10% vocabulary to express, but he must master almost 100% phonetics. (A. C. Gimson, 1980) Correct pronunciation helps students remember and accumulate vocabulary and build benign vocabulary acquisition mechanisms. Active speech learning can improve the beauty and accuracy of spoken language. Acquisition of speech recognition knowledge and skills will help and ensure smooth communication. English intonation has always been one of the difficult points in which learners learn English pronunciation. (Banjo, 1979). Even the learner’s monophonic pronunciation is very standard, intonation is still a obstacle to eliminate foreign accents. (Cruz-Ferreira, 1989) Therefore, since the late 1980s, the key of phonetic research and language teaching begins to transfer from the segmentals to suprasegmentals. Along with the development of computer technology, it has become an important area for researchers to explore intonation by the use of speech visualization. This study will focus on the Acquisition of English Intonation by Native Chinese speakers with praat analysis. Praat provides a contrast of different pronunciation, allowing researchers and learners to observe the "deviation" of the pronunciation from the native speaker’s to record, practice and analyse. It has also been found that speakers who receive instruction that emphasizes prosodic functions such as rhythm, intonation, and stress can shift their learning to spontaneous production more effectively than those who receive instruction only with segmented content. (Derwing, Munro, & Wiebe 1998) Second language learner errors have been shown to be due to the ability of native speakers (Broselow 1987). In this paper, some of the specific issues in English intonation are analysed to point out similarities and differences that should be emphasized by Chinese English learners. 2. Literature Review Along with the growth in research on speech learning, there have been significant technological advances in speech analysis software. Although the accessibility and functionality of early speech analysis systems was limited, recent developments such as Praat are powerful (Boersma and Weenink 2011), free to download, and very versatile. Researchers examined the potential benefits of visually analyzing the learner's spontaneous speech and comparing it to native speakers (Abberton and Fourcin 1975). Specifically, L2 learners, native French speakers, were determined to be more accurate at the segmental level after the training period. Similarly, Saito (2007), using analysis software, showed an improvement in the accuracy of Japanese vowel pronunciation in Japanese, presented as a visual image of his work as a native English speaker. Lord(2005) presents one such effort and examines the improvement of Spanish-speaking learners in courses in advanced-level phonetics courses. More recently, Lord (2010) has used speech analysis software as a component tutor in the interaction between audio tutoring and immersive experiences in Spanish vocabulary. However, it should be noted that these approaches to speech analysis software are only investigating the impact of more language instruction. Intonation is a fundamental property of spoken language. This is a supersegmental feature of the language. Would be said to be a feature not on a segment vowels and consonants only, but above them, created by adding and connecting several single segments. "linguists have found many different ways to characterize tones. This is also called pitch contour or pitch motion." In some languages, such as Persian, intonation is translated as a sentence Music (Samareh, 1984). The most important part is the pitch. It is related to the frequency of vocal fold vibration. The faster the vocal fold vibration, the higher the pitch. The meaning can be distinguished at the segment level. It is a component of stress and forms the intonation of connected speech. Only in very rare situations will we talk at a fixed, constant pitch. In the normal story, the pitch of the voice is constantly changing (Roach, 2000). Thus, tone is a hyper-segmental feature of all languages, mainly pitch oscillations and prominence for a series of sounds. For the purposes of this paper, the following broad definitions of intonation are adopted: pitch levels, pitch movements and systematic changes in prominence. Intonation can split into three parts: (1) Tonic: Tonic or nuclear placement. (2). tonality: speech into tone groups or intonation phrases (3) Tone: different melodic Available contours. These options are also called "three Ts" (Wells, 2006). In China, pronunciation has traditionally played a secondary role than grammar, vocabulary, and productive language skills. English pronunciation guidance for students are neglected. Pronunciation teaching focus on teaching students to pronounce correct word. In recent years, the focus has been emphasized for supersegmental features such as stress and intonation. The focus of this empirical research is on some of these supersegmental features, called intonation. Chinese English teachers is actually a problem for Chinese EFL learners, but rarely addressed. As for teachers who teach English in China, the reality is that teachers do not have enough knowledge of intonation. In addition, during interviews, some Chinese teachers said they did not like teaching intonation because they did not think they would be a good model. In addition, English and Chinese belong to different language systems. This difference may be a potential barrier for Chinese EFL learners in studying English intonation. Differences in the use of intonation in English and Chinese and lack of trust among teachers are factors that cause English intonation to not be properly taught in China. As a result, intonation errors are widespread in Chinese EFL learners. The computer assisted learning makes a great improvement of the spoken English training and acquisition. In recent years, Chinese researchers and English teachers have proposed many ways to teach English pronunciation. For example, based on Spoken Language Law, English speech education mainly uses correction, imitation, and comparison methods to develop the habit of pronounce English words and sentences. With a communicative approach, teachers have begun to focus on teaching the intonation of English (Gao, 2011). Overall, researchers and teachers have made significant efforts to improve the speech abilities of Chinese EFL learners. However, according to a study by Hu (2013), Chinese EFL learners still speak English with a strong Chinese accent, even in dialects. The current state of English voice learning can be summarized as follows. A common problem for Chinese EFL learners with English rhythms is that they do not distinguish between stressed and unstressed syllables. Generally speaking, students tend to adopt lower intonation to read all declarations and use higher intonation in questions. Flat intonation and rising intonation in sentences are rarely adopted. As a result, intonation becomes dull and lacks intonation changes. One of the important speech phenomena of English pronunciation in modern speech is that the speech changes depending on the speech context. Chinese EFL learners rarely use these sound variation rules. As a result, they pronounce all words clearly and completely, making them difficult for native English speakers to understand. 3. Praat Praat is a great audio software that can visualize EFL pronunciation tests and evaluations by providing a visual method of conceptualizing pronunciation. Illustrations for better analysis of sound waves, extracting specific values such as formants, fundamental frequency, intensity, pitch and duration, etc. for general evaluation of segmentation, stress and intonation. Praat is co-developed set by Paul Boursma and David Weenink of the University of Amsterdam, now available free of charge online. (Zhen Li, 2014) Praat is free, downloadable software, available for PC, Mac and Linux (Boersma and Weenink 2011). It can record a voice signal and convert it to a sound wave pattern and generate the corresponding parameter diagram. Speech evaluation usually involves three items segmentation, stress, and intonation. Among them, stress and intonation are used to examine the phonemic properties of supersegments. More specifically, this sort of speech analysis software has the potential to play a significant role in the development of self-monitoring and comparing one's own productions with productions of native speakers. The reliability and validity of the evaluation is thus improved, and a new idea for self-study and self-evaluation is therefore developed. There is simply no better package for linguists to use in analyzing speech than praat. Many existing software documentation is simply software documentation and is not designed to assist linguists. The Software instructions provide a way for linguists to perform some of the tasks that they want to do but without a strong phonetic or programming background. Praat also offers some of the more complex measures and tricks. How to make voice measurements and analysis by Praat? You could download and select a sound to open it in the editor window, then work on the plateau waveform and spectrogram. Immediately, an editor window will appear, showing the sound waveform. If the sound is short enough, a broadband spectrogram showing the spectral energy of the sound over time is displayed. You may also see a series of red dots representing formants, a blue line representing speaker pitch, and a yellow line representing intensity. If you see two waveforms in the window, you are opening a stereo sound file. The two displayed channels are combined and displayed. The most important settings are the window length and display range. The view range controls the amount of spectrum displayed. For speech, you may be interested in the range 0-5000 or 6000 Hz, but if you are looking for fricatives, you may want to be as high as 15,000 Hz. In any case, this is a way to set which part of the spectrum you are interested in. For example, what is the reliable way to examine and measure F0 / Pitch in Praat? The most reliable way to get the correct F0 in one cycle is to open the file in an editor window and then zoom in to a point in the sound file where you can see individual cycles; Select exactly one complete cycle as possible and give Praat a duration in seconds; Praat calculates the sound frequency in Hertz in the top bar and gives the result. The complicated data analysis will involve the programme to get more reliable data analysis. 4. Research Methodology This study will focus on the comparison and analysis of the Chinese English learners and the native speakers’ speech pitch to find the characters of intonation acquisition of Chinese native English learners. The research questions are as follows: 1. What differences are there between the pitch of the native speaker and Chinese English learner? 2. What’s the difference of the pitch between the different levels of Chinese English proficiency learners? 3. How to deal with the instrumental methods of studying intonation: measurement, modelling and manipulation 4. What are the different quantitative models of pitch? Research Material is taken from English Fun Duding, which is regarded as one of the most popular spoken English learning platform. English Fun Dubbing is an English learning software developed by Hangzhou Feishu Technology Co., Ltd. Dubbing short videos of 1-2 minutes make learning English fun. The APP updates the latest video resources such as American dramas, anime, songs and more. English learner and English enthusiasts can get rid of boring supportive learning methods, and freely choose to imitate and follow their favourite videos and fall in love with English learning. It has won the Baidu Golden Bear Palm Recommendation Award and has been recommended as a highquality boutique application by the App Store. English Dubbing Practice English Dubbing with Online Dubbing of English Videos learners could select the dubbing of the videos they are interested in and learn the spoken English. The software lists each sentence in the video separately. The level of English is high and the effect of the dubbing is highly appreciated. English learners can learn English by dubbing, especially by pronunciation and intonation. This allows learners to improve unconsciously spoken English while dubbing. The sample of native speakers selected is the Queen's Christmas message for 2015. The Queen's English, also known as the Received pronunciation, is a famous English accent spoken by British royals and other members. It is an accent that attracts many non-native speakers, and many English students try to speak with elegant accents. Some natives try to copy the accents to make them more native or for fun. The study participants are three English learners of the APP English Fun Duding with different levels of English proficiency. Three random recordings from the whole are selected as targets for the next analysis. Two of them are female and one is male. Selected samples are transcribed in intonation notations, nonality and Tonicity nucleus. Participant recordings will be transcribed in the same way. Thus, intonation issues can be found in the comparison between native speakers and participants. The materials are provided as follows: 1. At this time of year, few sights evoke more feelings of cheer and goodwill than the twinkling lights of a Christmas tree. 2. The popularity of a tree at Christmas is due in part to my great-great grandparents, Queen Victoria and Prince Albert. 3. After this touching picture was published, many families wanted a Christmas tree of their own, and the custom soon spread. The Research tool is the free software Praat. Participant recordings were extracted and analyzed using Praat (Boersma & Weenink, 2014). Praat is a very flexible tool for performing speech analysis. Provides a wide range of standard and non-standard procedures, such as spectroscopy, articulation, and neural networks. In this study, Praat will be used to analyse participants' tone and tone performance. I could clearly see the comparison between the native speaker and the Chinese English leaner. In Praat, the waveform shows the pressure over time. The numbers on the left indicate the amount of amplitude. The syllable segmentation of the Praat text grid allows you to read the range of pitch carefully. This is based on the claim of tonicity in the intonation phrase along with reading the context. Below is an example of transcription Pratt: (Figure 2) 5. Data Analysis This part presents data relevant to this study. Then the relevant basic intonation facts are explained and explained by contrasting data between native speakers and Chinese English learners. Content words are nouns, adjectives, most verbs, and most adverbs. Example: desk, book, remember, green, good luck. Function words, on the other hand, pronouns, prepositions, articles, auxiliary verbs, and legal verbs are as follows. Words whose meaning needs to be explained in grammar. In particular, the nucleus is usually a personal pronoun, preposition, or auxiliary verb. These words are core only when special circumstances apply. If IP ends with a personal pronoun or preposition. The nucleus usually proceeds faster.In addition to word placement and category, the location of the nucleus is strongly influenced by information status of the word, that is, if the utterance contains old or new information. The general rule is to add a new accent Information, not old information. Therefore, nuclear locations usually signal the end of new information on intonation phrases. For example, because the last content word in the IP is important for meaning and accented, it will take the nucleus. And this is called "neutral" or "unmarked" tonicity. In contrast, where contrast and freshness are involved, there are also "marked tonicities" where the nucleus does not fall into the last content word (Ladd, 1996; Wells, 2006). Neutral tonicity error consists of assigning the nucleus of the tone unit to other words. The conditions of the application are always more than the last vocabulary of the tone unit.This means that function words at the end of the IP are incorrectly placed by the learner as nucleus. The wrong nucleus is marked with an asterisk just before the nucleus. In the meantime, Praat photos will be provided for both native speakers and Chinese EFL learners. Phonologically we accent syllables by giving a noticeable change in pitch, pitch movement, or onset of pitch movement. The nucleus is more emphasized than the other emphasized syllables in the IP, and its pitch movement is also considered to be the most prominent. Thus, pitch contours provide one method of tonicity measurement, and their Praat photographs support the tonicity claim in the intonation phrase of this study. The following examples are all from the material assigned in this study. (1) At this time of year│few sights evoke │more feelings of cheer and goodwill │than the twinkling lights │ of *a Christmas tree║ (by participants) At this time of year│few sights evoke more feelings of cheer and goodwill │than the twinkling lights of a Christmas tree║ (by native speaker) (2) The popularity │*of a tree at Christmas│ is due in part to my great-great Grandparents │ Queen Victoria and Prince Albert║ (by participants) The popularity of a tree at Christmas │is due in part to my great-great grandparents│ Queen Victoria and Prince Albert║ (by native speaker) (3) After this touching picture │ was published, *many families wanted a Christmas tree │ of their own, and the custom soon spread║(by participants) After this touching picture was published │ many families wanted a Christmas tree of their own and the custom soon spread║(by native speaker) English native speakers do not accent articles or pronouns, if you can't find newness to contradiction. But Chinese learn ers tend to emphasize them. The Praat picture is recorded from the Chinese learner. The study find that the pitch movement of the article a is the most prominent. Christmas, however, does not have that pitch movement or change. It shows that the Chinese learner put the nucleus on the pronoun a rather than the noun Christmas. Meanwhile, the picture uttered by the native speaker presents that the article a does not own much changes in pitch value. Christmas is actually the nucleus because it begins the rise movement of pitch. Neutral tonicity errors or problems among Chinese learners are presented in the data. This can be summarized that Chinese learners tend to place tonicity on the last word of the IP, regardless of word category. Several observations can be made regarding the qualitative aspects of neutral tonicity errors. First, the core placement of English function words such as personal pronouns, articles, conjunctions, and auxiliary verbs in Chinese EFL learners is a type of error that can be classified as interference errors. The reason is that a neutral nucleus is not a functional word even if it ends with IP. On the contrary, in the Chinese language system, the nucleus is most often in the word category, the last word, regardless of the content word or the function word. Therefore, EFL learners usually need to learn not to place nucleus on English functional words. Second, incorrect assignment of nucleus to articles or conjunctions constitutes not only neutral tonic errors, but also tonal errors. Respectively we can conclude that there is a close relationship between tonicity and tonality. Third, neutral tonicity errors for compounds are classified as interference errors too. In English, generally speaking, the first element of a compound is linguistically more important than the second. As a result, The nucleus, the first element of the compound is where the nucleus goes exactly. In summary, Chinese EFL learners must learn to de-accent not only final function words but also content words with old information, whether final in the tone unit or not; to place the nucleus on a word containing contrastive information, whether it is a content or a function word and whether or not it occurs in final position. 6. Conclusion In this study, tonicity is further divided into neutral tonicity and pronounced tonicity. However, whether tonicity is natural or prominent, a common problem for EFL learners is that the last word of the intonation phrase is always nucleus, regardless of the location of the word, the category of the word, or the state of the information. In terms of neutral tonicity, Chinese EFL learners incorrectly assign nucleus to articles, personal pronouns, conjunctions, and auxiliary verbs. There should be no neutral tonicity errors. In fact, Chinese learners have found that the general rule of marked tonicity is to place in the last content word. Incorrect assignment of nucleus to articles or conjunctions constitutes not only neutral tonic errors, but also tonal errors. Chinese EFL speaker places an IP boundary immediately after an article or conjunction, because the article or conjunction is the last word. Chinese EFL learners also have problems with correctly placing the nucleus in compound words, which also belongs to the neutral tonicity problem. In this study, Chinese learners cannot give compound words the right nucleus. This could be the further study topic of the study. In terms of marked tonicity, Chinese EFL learners ignore the information of the status of words. They are not aware that new and contrasting information is always important for the marked tonicity. Reference: 1. Abberton, Evelyn, and Adrian Fourcin. (1975). "Visual Feedback and the Acquisition of Foundations of Language Development: A Multidisciplinary Approach. Ed. Eric H. Lenneberg Elizabeth Lenneberg. Vol. 2. New York: Academic. 2. Zhen Li . “Visualization of Pronunciation Testing and Evaluation By Praat.” [J]. Foreign Language,2014,30(06):98-103. 3. Banjo, A. Beyond Intelligibility in Nigerian English [J] Varieties and Functions of English in Nigeria, E. Ubahakwe (ed.), 1979. 7–13. Ibadan: African Universities Press. 4. Boersma, Paul, and David Weenink. (2011). Praat: Doing Phonetics by Computer Version 5.2.16. Web. 20 Feb. 2011. 5. Broselow, E. Non-obvious transfer: On predicting epenthesis errors. In G.Ioup and S. Weinberg (Eds.). Interlanguage phonology: The acquisition of a second language system. Cambridge: Newbury House Publishers.1987 6. Cruz-Ferreira, M. Non-native comprehension of intonation patterns in Portuguese and in English[D]. Unpublished Ph.D. Thesis, University of Manchester,1983. 7. Derwing, T.M., M.J. Munro & G. Wiebe. 1998. Evidence in Favour of a Broad Framework for pronunciation Instruction. Language Learning, 48, 3: 393-410 8. Gao, Lin. (2011). A review of English phonetic acquisition of Chinese learners, Journal of Donghua 11.1, 31-35. 9. Gimson，A.C. . “ An Introduction to the Pronunciation of English “ [M]. London: Edward Arnold，1980. 10. Hu, Fanghui. (2013). Analysis on non-English majors’ pronunciation learning and approaches to improve their pronunciation proficiency. Overseas English 252.21, 260-261. 11. Hu, Fanghui. (2017), A Study on Chinese EFL Learning of English Pronunciation from the Perspective of Aesthetic Linguistics, Theory and Practice in Language Studies, Vol. 7, No. 7, pp. 579-584, July 2017 12. Ladd, D. R. Intonational Phonology [M]. Cambridge: Cambridge University Press, 1996. 13. Lord, Gillian. (2005). "Can We Teach Foreign Language Pronunciation? The Effects of a Phonetics Class on Second Language Pronunciation." Flispania 88.3: 557-67. Print. (2008). "Podcasting Communities and Second Language Pronunciation." Foreign Language Annals 41: 364-79. Print. -. (2010). "The Combined Effects of Instruction and Immersion on Second Language Pronunciation." Foreign Language Annals 43.4: 488-503. 14. Olson, Daniel J. “Phonetics and Technology in the Classroom: A Practical Approach to Using Speech Analysis Software in Second-Language Pronunciation Instruction.” Hispania, vol. 97, no. 1, 2014, pp. 47–68. JSTOR 15. Saito, Kazuya. (2007). "The Influence of Explicit Phonetic Instruction on Pronunciation Teaching in EFL Settings: The Case of English Vowels and Japanese Learners of English." The Linguistics Journal 3.3: 16-40. Print. 16. Wells, J. C. English Intonation: An Introduction [M]. Cambridge: Cambridge University Press, 2006.