일본어 합성기에서 악센트 정보가 결합된 발음기호를 이용한 Break 예측 방법 = Break Predicting Methods Using Phonetic Symbols Combined with Accents Information in a Japanese Speech Synthesizer

다국어 초록 (Multilingual Abstract)

Japanese is a language having intonations, which are indicated by the relative differences in pitch heights and the accentual phrases (APs) are placed according to the changes of the accents while a break occurs on a boundary of the APs. Although a break can be predicted by using J-ToBI, which is a rule-based or statistical approach, it is very difficult to predict a break exactly due to the flexibility. Therefore, in this paper, a method which can enhance the quality of synthesized speech by reducing the errors in predicting break indices (BI), are proposed. The method is to use a new definition for the phonetic symbols, which combine the phonetic values of Japanese words with the accents information. Since a stream of defined phonetic symbols includes the information on the changes in intonations, the BI can be easily predicted by dividing the intonation phrase (IP) into several APs. As a result of an experiment, the accuracy of break generations was 98 % and the proposed method contributed itself to enhance the naturalness of synthesized speeches.

참고문헌 (Reference)

1 전성용, "일본어의 발음과 악센트, 1st Ed" Japanese Technical Publishing Company 5-11, 2002

2 "http://www.oki.com/jp/RDG/JIS/oto/tts/"

3 "http://www.cjk.org/cjk/"

4 H. Kawai, "Ximera: A new TTS from ATR based on corpus-based technologies" 179-184, 2004

5 R. E. Donovan, "Trainable speech synthesis" Engineering Department 1-28, 1996

6 S. Narayanan, "Text to speech synthesis: New paradigms and advances, 1st Ed" Prentice Hall Professional Technical Reference 155-173, 2004

7 "Speech synthesis system performance evaluation methods" 42-45, 2003

8 H. Fujisaki, "Prosodic parameterization of spoken Japanese based on a model of the generation process of F0 contours" 2439-2442, 1996

9 A. Conkie, "Preselection of candidate units in a unit selection-based text-to-speech synthesis system" 3 : 314-317, 2000

10 J. Venditti, "Japanese ToBI labeling guidelines" 127-162, 1997

11 T. Kazuyo, "JEIDA standard of symbols for Japanese text-to-speech synthesizers" 27-32, 2000

12 Y. Kitahara, "Development of high quality and intelligent speech synthesis technology" 88 (88): 60-65, 2006

13 D. S. Na, "A preselection method using accentual phrase matching in unit selection-based Japanese text to speech" IX : 124-, 2006