Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

Relationship Among Speaking Style, Inter-Phoneme's Distance and Speech Recognition Performance

Kazumasa Yamamoto (1), Seiichi Nakagawa (2)

(1) Dept. of Electrical & Electronic Eng., Faculty of Engineering, Shinshu University, Nagano, Japan
(2) Department of Information and Computer Sciences, Toyohashi University of Technology, Japan

There is a limit of recognition performance for dialogue speech using acoustic models built only with read speech, because various acoustic and linguistic phenomena, which reflect the characteristics of spontaneous speech, are observed in the dialogue speech. In this paper, we inves- tigated the di erences of acoustic properties which cause the limit among isolated words, read speech and spontaneous speech. Firstly, the dialogue speech was compared with the read speech through acoustic analyses. Next, the acoustic models were separately built with each of the speech databases. The recognition performance was experimentally evaluated using the acoustic models and the relations of the di erences of the performance to those of the acoustic features observed in the analyses were investigated quantitatively. The e ectiveness of speaker adaptation was also investigated in the same manner.

Full Paper

Bibliographic reference.  Yamamoto, Kazumasa / Nakagawa, Seiichi (2000): "Relationship among speaking style, inter-phoneme's distance and speech recognition performance", In ICSLP-2000, vol.2, 859-862.