Signal enhancement device, method thereof, program, and recording medium

作者: Tomohiro Nakatani , Masato Miyoshi , Takuya Yoshioka

DOI:

关键词:

摘要: The initial values of parameter estimates are set, including reverberation estimates, which includes a regression coefficient used in linear convolutional operation for calculating an estimated value included observed signal, source prediction and residual power that identify the spectrum noise include estimates. Then, maximum likelihood estimation is to alternately repeat processing updating at least one until predetermined termination condition satisfied.

参考文章(11)
Tetsuya Takiguchi, Masafumi Nishimura, Signal enhancement and speech recognition ,(2005)
Jae Lim, A. Oppenheim, All-pole modeling of degraded speech IEEE Transactions on Acoustics, Speech, and Signal Processing. ,vol. 26, pp. 197- 210 ,(1978) , 10.1109/TASSP.1978.1163086
Takuya Yoshioka, Takafumi Hikichi, Masato Miyoshi, Dereverberation by using time-variant nature of speech production system EURASIP Journal on Advances in Signal Processing. ,vol. 2007, pp. 6- 6 ,(2007) , 10.1155/2007/65698
Marc Delcroix, Takafumi Hikichi, Masato Miyoshi, Dereverberation and Denoising Using Multichannel Linear Prediction IEEE Transactions on Audio, Speech, and Language Processing. ,vol. 15, pp. 1791- 1801 ,(2007) , 10.1109/TASL.2007.899286
M. Delcroix, T. Hikichi, M. Miyoshi, On the Use of Lime Dereverberation Algorithm in an Acoustic Environment With a Noise Source international conference on acoustics, speech, and signal processing. ,vol. 1, pp. 825- 828 ,(2006) , 10.1109/ICASSP.2006.1660148
雅清 藤本, Masakiyo Fujimoto, Satoru Nakamura, 哲 中村, Disturbance component suppressing device, computer program, and speech recognition system ,(2005)