Spatial Loss for Unsupervised Multi-channel Source Separation
Spatial Loss for Unsupervised Multi-channel Source Separation
复制标题
无监督多通道源分离的空间损失
DOI：
10.21437/interspeech.2022-274
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Robin Scheibler
中科院分区：
文献类型：
--
作者：
Kohei Saijo;Robin Scheibler
我们提出了一个空间损失的无监督多通道源分离。所提出的损失利用到达方向（DOA）和波束形成的对偶性：导向和波束形成向量应该针对目标源对齐，但对于干扰源正交。空间损失导致了经典DOA估计器和神经元分离器的混合和去混合系统之间的一致性。在提出的损失下，我们基于最小方差无失真响应（MVDR）波束成形和独立向量分析（IVA）训练神经分离器。我们还研究了结合我们的空间损失和信号损失，它使用盲源分离的输出作为参考的有效性。我们对合成和记录（LibriCSS）混合物评估了我们提出的方法。我们发现空间损失对于训练基于IVA的分隔符是最有效的。对于神经MVDR波束形成器，当与信号丢失相结合时，它表现最好。在合成混合物上，所提出的无监督损失导致在字错误率方面与监督损失相同的性能。在LibriCSS上，我们在没有任何标记训练数据的情况下获得了接近最先进的性能。
We propose a spatial loss for unsupervised multi-channel source separation. The proposed loss exploits the duality of direction of arrival (DOA) and beamforming: the steering and beamforming vectors should be aligned for the target source, but orthogonal for interfering ones. The spatial loss encour-ages consistency between the mixing and demixing systems from a classic DOA estimator and a neural separator, respec-tively. With the proposed loss, we train the neural separators based on minimum variance distortionless response (MVDR) beamforming and independent vector analysis (IVA). We also investigate the effectiveness of combining our spatial loss and a signal loss , which uses the outputs of blind source separation as the references. We evaluate our proposed method on synthetic and recorded (LibriCSS) mixtures. We find that the spatial loss is most effective to train IVA-based separators. For the neural MVDR beamformer, it performs best when combined with a signal loss. On synthetic mixtures, the proposed unsupervised loss leads to the same performance as a supervised loss in terms of word error rate. On LibriCSS, we obtain close to state-of-the-art performance without any labeled training data.