改进的稀疏字典学习单通道语音增强算法

An Improved Monaural Speech Enhancement Algorithm Based on Sparse Dictionary Learning

基于K奇异值分解字典学习方法及其非负约束下的修改算法,本文提出一种改进的单通道语音增强算法。该算法将噪声划分为结构化噪声和非结构化噪声两部分。首先通过稀疏字典学习的方法对结构化噪声进行建模,训练出噪声字典;然后,使用所得噪声字典去除带噪语音中的结构化噪声;最后,采用过完备字典和稀疏表示的方法对纯净语音进行提取,去除非结构化噪声。实验结果表明,在平稳或非平稳噪声环境下,本文算法均能有效去除加性噪声,性能优于多带谱减法和基于非负稀疏编码的增强算法。

This paper applies the K-Singular Value Decomposition method and its non-negative variant to enhance the contaminated speech. In the proposed approach, noise is categorized as structured and unstructured noise. Firstly, the noise dictionary is learned from a training noise database. Then, we remove the structured noise iteratively by using the noise dictionary. Finally, the approach adopts sparse and redundant representations over trained dictionary to separate the clean speech from the unstructured noise. Extensive experimental results show that the enhancement method proposed out- performs state-of-the-art methods like muhi-band spectral subtraction and the non-negative sparse coding based noise reduc- tion algorithm.