이층위 신경망 모형을 사용한 포먼트 궤적 기반 모음 유형 분류
Two-layer neural network-based vowel classification experiments using formant trajectory
0
조회
0
다운로드
9
인용
0
참고문헌
키워드
초록
The aim of this paper is to classify vowel types attested in the TIMIT corpus of American English using a two-layer neural net with the input feature of dynamic formant trajectories. Unlike previous studies which used a single (near the middle) or two static points (near the beginning and ending points within the duration of a vowel), the current study tried to incorporate feature values related to formant trajectories by sampling F1 and F2 values at every 10% of the duration across vowels produced in a connected speech corpus. Experimental results showed that the 20 vowel categories annotated in the TIMIT corpus could be accurately predicted at a rate of about 60.1% with a model of a two-layer neural net. The results confirmed that trajectory-based features were a better predictor then formant values obtained from one or two points. Even though the accuracy reported in the paper is comparable to or better than the accuracy rates reported in previous literature, further research needs to be done which can incorporate contextual information.
Abstract
The aim of this paper is to classify vowel types attested in the TIMIT corpus of American English using a two-layer neural net with the input feature of dynamic formant trajectories. Unlike previous studies which used a single (near the middle) or two static points (near the beginning and ending points within the duration of a vowel), the current study tried to incorporate feature values related to formant trajectories by sampling F1 and F2 values at every 10% of the duration across vowels produced in a connected speech corpus. Experimental results showed that the 20 vowel categories annotated in the TIMIT corpus could be accurately predicted at a rate of about 60.1% with a model of a two-layer neural net. The results confirmed that trajectory-based features were a better predictor then formant values obtained from one or two points. Even though the accuracy reported in the paper is comparable to or better than the accuracy rates reported in previous literature, further research needs to be done which can incorporate contextual information.
참고문헌 (35)
- 윤태진, "성대진동 및 성별이 미국영어 마찰음에 미치는 효과에 관한 코퍼스 기반 연구", 말소리와 음성과학, 10(2), pp.7~14 (2018), 한국음성학회[DOI]
- ADANK, PATTI., "Vowel Normalization: A Perceptual-acoustic Study of Dutch Vowels", PhD, pp.2003
- BANG, HYE-YOUNG, "The emergence, progress, and impact of sound change in progress in Seoul Korean : implications for mechanisms of tonogenesis", Journal of Phonetics, 66, pp.120~144 (2018)
- BISHOP, CHRISTOPHER., "Neural Networks for Pattern Recognition" (1995), Oxford University Press
- BOERSMA, PAUL, "Praat: Doing phonetics by computer (Version 6.0.25)", ~ (2018)
- BOWIE, DAVID, "Acoustic Characteristics of Utah’s card-cord merger", American Speech, 83, pp.35~61 (2008)
- BYRD, DANI, "Relations of sex and dialect to reduction", Speech Communication, 15, pp.39~54 (1994)
- Chung, Hyunju, "Vowel Formant Trajectory Patterns for Shared Vowels of American English and Korean", 말소리와 음성과학, 2(4), pp.67~74 (2010), 한국음성학회
- COLE RONALD, "Perceptual studies on vowels excised from continuous speech", Proceedings of ICSLP, pp.1091~1094 (1992)
- COLE RONALD, "The contribution of consonants versus vowels to word recognition in fluent speech", Proceedings of the International Conference on Acoustics, Speech, and Signal Processing(ICASSP’96), pp.853~856 (1996)
- FOGERTY, DAINEL, "Perceptual contributions of the consonant-vowel boundary to sentence intelligibility", Journal of the Acoustical Society of America, 126, pp.847~857 (2009)
- GOKI, SAITO., "Deep Learning from Scratch" (2016), O’Reilly Japan
- HILLENBRAND, JAMES, "Identification of resynthesized", hVd, Journal of the Acoustical Society of America, pp.1999, 105[DOI]
- HILLENBRAND, JAMES, "Acoustic characteristics of American English vowels", Journal of Acoustical Society of America, 97(5), pp.3099~3111 (1995)
- 홍순현, "Pattern recognition modeling of American English vowel identification by four different identification-proficiency levels of Korean listeners", 음성음운형태론연구, 22(1), pp.147~175 (2016), 한국음운론학회[DOI]
- JOOS, MARTIN, "Acoustic Phonetics", Language, 24(2), pp.5~136 (1948)
- KEWLEY-PORT, DIANE, "Contribution of consonant versus vowel information to sentence intelligibility for young normal-hearing and elderly hearing-impaired listeners", Journal of the Acoustical Society of America, 122, pp.2365~2375 (2007)
- KLATT, DENNIS, "Linguistic uses of segmental duration in English : acoustic and perceptual evidence", Journal of the Acoustical Society of America, 59, pp.1208~1221 (1976)
- KLUENDER, KEITH, "Vowel-length differences before voiced and voiceless consonants : an auditory explanation", Journal of Phonetics, 16, pp.153~169 (1988)
- LADEFOGED, PETER., "A Course in Phonetics" (2011), Cengage Learning
- LEHISTE, ILSE., "Suprasegmentals" (1970), MIT Press
- MENG HELEN M., "Signal representation comparison for phonetic classification", Proceedings of ICASSP (1991)
- MORRISON, GEOFFREY S., "Vowel Inherent Spectral Change" (2013), Springer
- NEAREY, TERRANCE M., "Modeling the role of inherent spectral change in vowel identification", Journal of the Acoustical Society of America, 80, pp.1297~1308 (1986)
- PETERSON, E. GORDON, "Control methods used in a study of the vowels", Journal of the Acoustical Society of America, 24, pp.175~184 (1952)
- RASHID, RARIQ., "Make Your Own Neural Network" (2016), CreateSpace Independent Platform
- STRANGE, WINFRED, "Dynamic specification of coarticulated German vowels : Perceptual and acoustical studies", Journal of the Acoustical Society of America, 104, pp.448~504 (1989)
- STRANGE, WINFRED, "Dynamic specification of coarticulated vowels", Journal of the Acoustical Society of America, 74, pp.695~705 (1983)
- TAUBERER, JOSHUA, "Intrinsic vowel duration and the post-vocalic voicing effect : some evidence from dialects of North American English", Proceedings of Interspeech, pp.2211~2214 (2009)
- WEENINK, DAVID, "Speaker-adaptive Vowel Identification", PhD, pp.2006
- WHALEN, DOUGLAS H., "The universality of intrinsic F0 of vowels", Journal of Phonetics, 23, pp.349~366 (1995)
- YAEGER, MALCAH., "Context Determined Variation in Montreal French Vowels", PhD, pp.1979
- ZAHORIAN, STEPHEN A., "Spectral-shape features versus formants as acoustic correlates for vowels", Journal of the Acoustical Society of America, 94, pp.1966~1982 (1993)
- ZUE, VICTOR, "Speech database development at MIT: TIMIT and beyond", Speech Communication, 9(4), pp.351~346 (1990)
- ZUE, VICTOR, "Transcription and alignment of the TIMIT database", Proceedings of the Second Meeting on Advanced Man- Machine Interface through Spoken Language, pp.11.1~11.10 (1988)