Variational Bayesian multi-channel robust NMF for human-voice enhancement with a deformable and partially-occluded microphone array
Variational Bayesian multi-channel robust NMF for human-voice enhancement with a deformable and partially-occluded microphone array
复制标题
DOI：
10.1109/eusipco.2016.7760402
复制
发表时间：
2016-11
期刊：
影响因子：
--
通讯作者：
Yoshiaki Bando;Katsutoshi Itoyama;M. Konyo;S. Tadokoro;K. Nakadai;Kazuyoshi Yoshii;HIroshi G. Okuno
中科院分区：
文献类型：
--
作者：
Yoshiaki Bando;Katsutoshi Itoyama;M. Konyo;S. Tadokoro;K. Nakadai;Kazuyoshi Yoshii;HIroshi G. Okuno
提出了一种适用于可变形部分遮挡麦克风阵列的人声增强方法。虽然分布在软管形救援机器人长身体上的麦克风阵列对于在倒塌的建筑物下寻找受害者至关重要，但麦克风阵列捕获的人类声音受到非静止致动器和摩擦噪声的污染。标准的盲源分离方法不能使用，因为相对麦克风位置随时间变化，其中一些偶尔会被碎石遮蔽。为了解决这些问题，我们开发了一种贝叶斯模型，该模型将多通道幅度谱图分成稀疏和低秩分量（人声和噪声），而不使用相位信息，这取决于阵列布局。以时变方式估计每个麦克风处的语音水平，以减少被遮蔽的麦克风的影响。使用3米软管形机器人与8个麦克风的实验表明，我们的方法优于传统的方法的信噪比为2.7 dB。
This paper presents a human-voice enhancement method for a deformable and partially-occluded microphone array. Although microphone arrays distributed on the long bodies of hose-shaped rescue robots are crucial for finding victims under collapsed buildings, human voices captured by a microphone array are contaminated by non-stationary actuator and friction noise. Standard blind source separation methods cannot be used because the relative microphone positions change over time and some of them are occasionally shaded by rubble. To solve these problems, we develop a Bayesian model that separates multichannel amplitude spectrograms into sparse and low-rank components (human voice and noise) without using phase information, which depends on the array layout. The voice level at each microphone is estimated in a time-varying manner for reducing the influence of the shaded microphones. Experiments using a 3-m hose-shaped robot with eight microphones show that our method outperforms conventional methods by the signal-to-noise ratio of 2.7 dB.