영어 강세 교정을 위한 주변 음 특징 차를 고려한 강조점 검출
Prominence Detection Using Feature Differences of Neighboring Syllables for English Speech Clinics
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Prominence of speech, which is often called ‘accent,’ affects the fluency of speaking American English greatly. In this paper, we present an accurate prominence detection method that can be utilized in computer-aided language learning (CALL) systems. We employed pitch movement, overall syllable energy, 300-2200 Hz band energy, syllable duration, and spectral and temporal correlation as features to model the prominence of speech. After the features for vowel syllables of speech were extracted, prominent syllables were classified by SVM (Support Vector Machine). To further improve accuracy, the differences in characteristics of neighboring syllables were added as additional features. We also applied a speech recognizer to extract more precise syllable boundaries. The performance of our prominence detector was measured based on the Intonational Variation in English (IViE) speech corpus. We obtained 84.9% accuracy which is about 10% higher than previous research.
Abstract
Prominence of speech, which is often called ‘accent,’ affects the fluency of speaking American English greatly. In this paper, we present an accurate prominence detection method that can be utilized in computer-aided language learning (CALL) systems. We employed pitch movement, overall syllable energy, 300-2200 Hz band energy, syllable duration, and spectral and temporal correlation as features to model the prominence of speech. After the features for vowel syllables of speech were extracted, prominent syllables were classified by SVM (Support Vector Machine). To further improve accuracy, the differences in characteristics of neighboring syllables were added as additional features. We also applied a speech recognizer to extract more precise syllable boundaries. The performance of our prominence detector was measured based on the Intonational Variation in English (IViE) speech corpus. We obtained 84.9% accuracy which is about 10% higher than previous research.
참고문헌 (13)
- Cook, A, "AAT-미국식 영어발음 집중훈련 워크북", pp.18~37 (2007), 윌북
- Tamburini, F, "An automatic system for detecting prosodic prominence in American English continuous speech", International Journal of Speech Technology, 8, pp.33~44 (2005)
- Kochanski, G, "Loudness predicts prominence: fundamental frequency lends little", J. Acoust. Soc. Amer, 118(2), pp.1038~1054 (2005)
- Wang, D, "An acoustic measure for word prominence in spontaneous speech in Audio, Speech, and Language Processing", IEEE, 15(2), pp.690~701 (2007)
- Bagshaw, P. C, "Automatic Prosodic Analysis for Computer-Aided Pronunciation Teaching", 박사, pp.1994
- Jenkin, K. L, "Development and comparison of three syllable stress classifiers", ICSLP `96 Proceedings, pp.1457~1460 (1996)
- Howitt, A. W, "Automatic Syllable Detection for Vowel Landmarks", 박사, pp.2000
- Xie, Z, "Robust acoustic-based syllable detection", ICSLP (2006)
- Taylor, P, "Analysis and synthesis of intonation using the tilt model", Journal of the Acoustical Society of America, 107, pp.1697~1714 (2000)
- Sluijter, A, "Acoustic correlates of linguistic stress and accent in Dutch and American English", ICSLP'96 Proceedings, pp.630~633 (1996)
- 김용대, "Comparison of Boosting and SVM", 한국데이터정보과학회지, 16(4), pp.999~1012 (2005), 한국데이터정보과학회
- "A practical guide to support vector classification"[URL]
- "LIBSVM"[URL]