The Role of Sound Changes in the Speech Recognition System: A Phonetic Analysis of the Final Nasal Shift in Mandarin

The Role of Sound Change in the Speech Recognition System: A Phonetic Analysis of the Final Nasal Shift in Mandarin 楊孝慈 James H. Yang 國立雲林科技大學應用外語系 Department of Applied Foreign Languages National Yunlin University of Science and Technology yanght@yuntech.edu.tw 摘要 過去十幾年來，電腦語言學家一直嘗試要設計一個能辨識標準語的語音系統，並能兼顧 個別使用者的語音差異。目前，一些語音辨識系統已能針對使用者的年齡和性別，處理 不同的音頻問題。然而，語音變化的問題還沒納入考慮。因此本文提議將語音變化的現 象，整合到語音辨識系統的設計。本文以中文的字尾鼻音變化為例，分析它的語音特質， 並討論它對語音辨識系統的影響，最後提出一個能提升辨識字尾鼻音的解決方案。 Abstract Over the past decade, computational linguists have been striving to design a speech recognition system that is able to identify standard speeches and to accommodate sound variables caused by different individual accents. Furthermore, some speech recognition programs have been able to learn and identify distinctive sound frequencies due to the user’s age and gender. Nevertheless, regular sound alterations that occur in different varieties of a language have never been seriously considered in the design of the speech recognition system. Accordingly, this study proposes to incorporate the socio-phonological information about regular sound modifications to enhance the performance of Automatic Speech Recognition. To illustrate this point, this study investigates and analyzes the acoustic variation of the syllable-final nasal shift from the velar to the dental, which has been discovered to be one of the distinctive sound features that makes the variety of Mandarin spoken in Taiwan differ from that spoken in China. Following the phonetic analysis, this study discusses the effect of the nasal merger on the development of phonology-dependent speech technologies. It concludes by proposing a preliminary resolution to the identification of syllable-final nasals for the design of Automatic Speech Recognition. 關鍵詞：中文語音學，語音變化，鼻音，語音辨識系統 Key words: Mandarin phonetics, sound change, nasal, speech recognition system 1. Introduction My motivation to explore the nasal merger of Mandarin spoken in Taiwan originates from an incident in my life. My brother’s first baby was born on August 31st, 1999. He gave his son a name called Geng-ren /kəŋ.ʐən/ 1 (耕仁, meaning “to cultivate benevolence”). Interestingly, I found, as a native speaker of Mandarin in Taiwan, that I would easily mispronounce his name as Gen-ren [kən.ʐən] (跟人, meaning “to follow people”), rather than its standard pronunciation. Since the name was subject to mispronunciation and misunderstanding, I suggested to my brother that he should change the name; hence, he later selected another name Jia-he (家和, meaning “harmony in the family”). Because of this interesting incident, I came to realize that many native speakers of Mandarin in Taiwan seem to merge the syllable-final velar nasal /ŋ/ with the dental nasal /n/, hence neutralizing such minimal pairs as geng /kəŋ/ (耕, “to cultivate”) and gen /kən/ (跟, “to follow”). To explore this possible sound change, I later conducted a speech production experiment, which is discussed in the subsequent section. 2. Speech Production Experiment To investigate the possible nasal merger observed above, I addressed three research questions: 1) Is the syllable-final nasal modification a free variation or a conditioned alteration? 2) Does it occur in Mandarin spoken in Taiwan, China, or both? 3) Is it an ongoing or complete sound change? To address these questions, I invited 30 native Mandarin speakers to participate in the speech production experiment, including 11 males and 19 females, who were students …

Content truncated to 500 words