Musical-noise-free blind speech extraction integrating microphone array and iterative spectral subtraction
Musical-noise-free blind speech extraction integrating microphone array and iterative spectral subtraction
复制标题
DOI：
10.1016/j.sigpro.2014.03.010
复制
发表时间：
2014-09
期刊：
影响因子：
--
通讯作者：
Ryoichi Miyazaki;H. Saruwatari;Satoshi Nakamura;K. Shikano;Kazunobu Kondo;J. Blanchette;M. Bouchard
中科院分区：
文献类型：
--
作者：
Ryoichi Miyazaki;H. Saruwatari;Satoshi Nakamura;K. Shikano;Kazunobu Kondo;J. Blanchette;M. Bouchard
在本文中，我们提出了一种使用麦克风阵列的无音乐噪声盲语音提取方法，适用于非平稳噪声。在我们之前的研究中，发现优化的迭代谱减法（SS）可以实现语音增强，几乎不会产生音乐噪声，但这种方法仅对平稳噪声有效。所提出的方法包括通过独立分量分析（ICA）或多通道维纳滤波等进行迭代盲动态噪声估计，以及通过修改的迭代SS进行无音乐噪声语音提取，其中将多个迭代SS应用于每个通道，同时保持动态噪声估计器重用的多通道属性。此外，关于所提出的方法，我们讨论了将 ICA 应用于 SS 非线性失真信号的合理性。通过模拟真实世界的免提语音通信系统的客观和主观评估，我们发现所提出的方法优于传统方法。
In this paper, we propose a musical-noise-free blind speech extraction method using a microphone array for application to nonstationary noise. In our previous study, it was found that optimized iterative spectral subtraction (SS) results in speech enhancement with almost no musical noise generation, but this method is valid only for stationary noise. The proposed method consists of iterative blind dynamic noise estimation by, e.g., independent component analysis (ICA) or multichannel Wiener filtering, and musical-noise-free speech extraction by modified iterative SS, where multiple iterative SS is applied to each channel while maintaining the multichannel property reused for the dynamic noise estimators. Also, in relation to the proposed method, we discuss the justification of applying ICA to signals nonlinearly distorted by SS. From objective and subjective evaluations simulating a real-world hands-free speech communication system, we reveal that the proposed method outperforms the conventional methods.