A solution to residual noise in speech denoising with sparse representation
A solution to residual noise in speech denoising with sparse representation
复制标题
DOI：
10.1109/icassp.2012.6288956
复制
发表时间：
2012-03
期刊：
影响因子：
--
通讯作者：
Yongjun He;Jiqing Han;Shiwen Deng;Tieran Zheng;Guibin Zheng
中科院分区：
文献类型：
--
作者：
Yongjun He;Jiqing Han;Shiwen Deng;Tieran Zheng;Guibin Zheng
稀疏表示作为一种很有前途的技术，在信号处理领域得到了广泛的研究。近年来，稀疏表示被广泛应用于噪声环境下的语音处理，但由于语音的特殊性，仍有许多问题需要解决。使用稀疏表示的语音去噪的一个假设是语音在字典上的表示是稀疏的，而噪声的表示是密集的。不幸的是，这种假设在语音去噪场景中不成立。我们发现许多噪音，例如，多路重合和白色噪声在用干净语音训练的字典上也是稀疏的，导致稀疏增强中的严重残余噪声。为了解决这个问题，我们提出了一种新的残余噪声降低（RNR）方法，首先找到的原子，代表噪声稀疏，然后忽略它们在重建的语音。实验结果表明，该方法可以有效地降低残余噪声。
As a promising technique, sparse representation has been extensively investigated in signal processing community. Recently, sparse representation is widely used for speech processing in noisy environments; however, many problems need to be solved because of the particularity of speech. One assumption for speech denoising with sparse representation is that the representation of speech over the dictionary is sparse, while that of the noise is dense. Unfortunately, this assumption is not sustained in speech denoising scenario. We find that many noises, e.g., the babble and white noises, are also sparse over the dictionary trained with clean speech, resulting in severe residual noise in sparse enhancement. To solve this problem, we propose a novel residual noise reduction (RNR) method which first finds out the atoms which represents the noise sparely, and then ignores them in the reconstruction of speech. Experimental results show that the proposed method can reduce residual noise substantially.