Noise reduction using connectionist models

Author

Tamura, Shin´Ichi ; Waibel, Alex

Author_Institution

ATR Interpreting Telephony Res. Lab., Osaka, Japan

fYear

1988

fDate

11-14 Apr 1988

Firstpage

553

Abstract

Using a back propagation network learning algorithm, a four-layered feed-forward network is trained on learning samples to realize a mapping from the set of noisy signals a set of noise-free signals. Computer experiments were carried out on 12 kHz sampled Japanese speech data, using stationary and nonstationary noise. The experiments showed that the network can indeed learn to perform noise reduction. Even for noisy speech signals that had not been part of the training data, the network successfully produced noise-suppressed output signals

Keywords

computerised signal processing; interference suppression; speech analysis and processing; 2 kHz; back propagation network learning algorithm; computer experiments; connectionist models; four-layered feed-forward network; mapping; noise reduction; noise-free signals; noise-suppressed output signals; noisy signals; noisy speech signals; nonstationary noise; sampled Japanese speech data; speech analysis; speech processing; stationary noise; training data; Computer architecture; Computer networks; Laboratories; Mathematical model; Noise reduction; Phase noise; Signal mapping; Speech enhancement; Speech processing; Telephony;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1988. ICASSP-88., 1988 International Conference on

Conference_Location

New York, NY

ISSN

1520-6149

Type

conf

DOI

10.1109/ICASSP.1988.196643

Filename

196643