关键词: 语音识别, 抗噪声, 人工神经网络, 谱压缩

Abstract:

Aimed at the problem of noise in application of speech recognition, a new method of robust feature extraction is presented. The speech was decomposed by wavelet transformation, and it was compressed by spectral compression scheme related to human hearing mask theory. Experimental results of MATLAB simulation showed that high recognition rate could be obtained by using of the new feature in noise environment. It can make the best of robust characteristic of wavelet and reduce the difference between training and recognition environment, so it is a robust feature.

Key words: speech recognition, robust, artificial neural networks, spectral compression