Speech enhancement with missing data techniques using recurrent neural networks

Author

Parveen, Shahla ; Green, Phil

Author_Institution

Dept. of Comput. Sci., Univ. of Sheffield, UK

Volume

1

fYear

2004

fDate

17-21 May 2004

Abstract

This paper presents an application of missing data techniques in speech enhancement. The enhancement system consists of two stages: the first stage uses a recurrent neural network, which is supplied with noisy speech and produces enhanced speech; whereas the second stage uses missing data techniques to further improve the quality of enhanced speech. The results suggest that combining missing data technique with RNN enhancement is an effective enhancement scheme resulting in a 16 dB background noise reduction for all input signal to noise ratio (SNR) conditions from -5 to 20 dB, improved spectral quality and robust automatic speech recognition performance.

Keywords

recurrent neural nets; signal denoising; spectral analysis; speech enhancement; speech recognition; uncertainty handling; RNN enhancement; automatic speech recognition; background noise reduction; missing data techniques; noisy speech; recurrent neural networks; robust performance; spectral quality; speech enhancement; speech quality; Acoustic noise; Artificial neural networks; Automatic speech recognition; Background noise; Frequency; Noise robustness; Noise shaping; Recurrent neural networks; Signal to noise ratio; Speech enhancement;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-8484-9

Type

conf

DOI

10.1109/ICASSP.2004.1326090

Filename

1326090