臺灣博碩士論文加值系統

透過網路學習英語已經成為另一種學習語言的新寵兒，藉由多媒體與大量的學習教材使得利用網路學習英語變得更活潑生動。此論文主要為開發一個結合聽力與口說的英語練習系統，由於在網路上的學習系統或教材主要是針對閱讀、撰寫與聽力的英語學習平台，較少有結合口說的英語練習系統，此論文鑑於此而結合聽力與口說的特色來輔助學習者學習英語。在聽力練習，結合聽寫和傳統的語言學習機的學習功能、例如設定A-B區間以達重覆聽的功能，及跟讀等有用的功能來輔助聽力練習。在口說練習中，強調英語中音調 (Intonation) 的重要性，利要音高追蹤找出相對的音高進而得到語調的曲線。另外此論文針對口說提供一個語音同步 (Speech Alignment) 的工具，利用兩個不同人對相同句子的錄音作時間點同步校正，而後利用左右聲道播放兩個不同聲音，學習者可以透過耳朵判識相異處，而改正個人發音。

With the rapid development of Internet and multimedia technologies, the E-learning/distance-learning systems have been developed rapidly in recent years. Language training is one of the popular domains in distance-learning applications. To date, however, most of the systems are designed for listening, writing and reading training. Few systems are designed for “speaking” training. The objective of this study is to design a web-based training system by integrating the speaking and listening. In listening, the dictation and the function of traditional language learning machine are main features in our work. In speaking, we emphasize the importance of intonation in one sentence, using pitch tracking algorithm to obtain the pitch contour. We provide another novel tool to correct learner’s pronunciation, using speech alignment to adjust two utterances which were spoken by different persons.

Abstract I
Contents II
List of Figures III
List of Tables IV
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Introduction 1
1.3 Listening 2
1.4 Speaking 3
1.5 Organization of this Thesis 4
Chapter 2 Related Work 5
Chapter 3 Speech Analysis 8
3.1 Short term features 8
3.1.1 Energy 9
3.1.2 Zero-crossing rate 10
3.2 Endpoint Detection 11
3.3 Pitch Tracking 13
3.3.1 Autocorrelation Function 13
3.3.2 Sum Magnitude Difference Function 15
3.3.3 Combination of SMDF and ACF 16
3.4 Speech Synchronization 17
Chapter 4 System Implementation 23
4.1 Features in listening 23
4.1.1 Dictation 23
4.1.2 Functions of traditional language learning machine 24
4.2 Features in speaking 25
4.2.1 The presentation of pitch contour 26
4.2.2 Speech synchronization 26
4.3 User Interface 27
Chapter 5 Conclusion 29
5.1 Conclusion 29
5.2 Future work 29
References 31

[Savog02] M.H. Savoji, “A robust algorithm for accurate endpoint of speech, “ Speech Commin, vol.8, pp. 45-60, 1989
[Kuo02] Kuo-Yu Liu, Natalius Huang, Bo-Hung Wu, Wei-Ta Chu, Herg-Yow Chen, “The WSML System: Web-based Synchronization Multimedia Lecture System“, Video Demonstrations of ACM Multimedia 2002, Juan-les-Pins, France, pp. 662-663, December, 2002
[Fujii00] Fujii, S., Iwata, J., Hattori, M., Iijima, M., Mizuno, T., “Web-CALL”: a language learning support system using Internet. Parallel and Distributed Systems: workshops, Seventh International Conference on, 2000, 4-7 Page(s): 326-331, July 2000
[Silk97] Silke Witt, Steve Young. Computer-assisted Pronunciatoin Teaching based on Automatic Speech Recognition In Proceedings of Language Teaching and Language Technology, Groningen, Netherlands, Swet and Zeitlinger, 1997
[Cucc97] Cucchiarini, C., Strik, H., Boves, L., Automatic evaluation of Dutch pronunciation by using speech recognition technology. Automatic Speech Recognition and Understanding, 1997. Proceedings, 1997 IEEE Workshopon, 14-17 Dec. Page(s): 622-629, 1997
[Ying 96] G.. S. Ying and L. H. Jamieson and C.D. Michell, “A probabillitic approach to AMDF pitch diction”, Spoken language, 1996. ICSLP 96. Proceedings, Fourth International Conference on Volume:2, 1996, Page(s): 1201-1204 vol.2
[Shing03] Jyh-Shing, Roger Jang, and Yung-Sen Jan “On the implementation of Melody recognition on 8bit and 16bit microcontrollers,” information, communication and signal processing 2003, Vol. 2, pp. 704-708.
[Rone97] O. Ronen, L. Neumeyer, and H. Franco. Automatic Detection Of Mispronunciation For language Instruction Proc. Of EUROSPEECH 97, pp. 645-648, Rhodes, 1997
[Herr99] Herron, D., Menzel W., Atwell E., Bisiani R., Daneluzzi F., Morton R., Schmidt J. A. “Automatic localization and dagnosis of pronunciation errors for second-language learners of English”, Eurospeech 99, Budapest, 5-9 September, v.2, p.855-858, 1999