A study on vowel behaviour and its description by a statistical model
作者： M. Di Benedetto
DOI: 10.1109/ICASSP.1985.1168076
关键词:
摘要: A complete statistical model for italian vowels is presented. The classification method contains the adaptive procedure which makes possible evolution of regions characteristic each vowel in F1-F2 plane and effectiveness has been verified a previous work [1]. In this paper, part algorithm described its theoretical detail justified on basis an extended experimentation. Results are given proving accuracy classifier segments independent context prosody dependent upon prosody.
sci-hub.se 参考文章(2)
M. Di Benedetto, A. Lanaro, How to avoid vowel normalization in identification of vowels in continuous speech international conference on acoustics, speech, and signal processing. ,vol. 9, pp. 300- 303 ,(1984) , 10.1109/ICASSP.1984.1172809
H. Wakita, Normalization of vowels by vocal-tract length and its application to vowel identification IEEE Transactions on Acoustics, Speech, and Signal Processing. ,vol. 25, pp. 183- 192 ,(1977) , 10.1109/TASSP.1977.1162929