Knowledge Commons of Institute of Automation,CAS
Gated Recurrent Fusion of Spatial and Spectral Features for Multi-channel Speech Separation with Deep Embedding Representations | |
Fan, Cunhang1,3; Tao, Jianhua1,2,3; Liu, Bin1; Yi, Jiangyan1; Wen, Zhengqi1 | |
2020-10 | |
会议名称 | Annual Conference of the International Speech Communication Association |
会议日期 | October 25–29, 2020 |
会议地点 | Shanghai, China |
摘要 | Multi-channel deep clustering (MDC) has acquired a good performance for speech separation. However, MDC only applies the spatial features as the additional information, which does not fuse them with the spectral features very well. So it is difficult to learn mutual relationship between spatial and spectral features. Besides, the training objective of MDC is defined at embedding vectors, rather than real separated sources, which may damage the separation performance. In this work, we deal with spatial and spectral features as two different modalities. We propose the gated recurrent fusion (GRF) method to adaptively select and fuse the relevant information from spectral and spatial features by making use of the gate and memory modules. In addition, to solve the training objective problem of MDC, the real separated sources are used as the training objectives. Specifically, we apply the deep clustering network to extract deep embedding features. Instead of using the unsupervised K-means clustering to estimate binary masks, another supervised network is utilized to learn soft masks from these deep embedding features. Our experiments are conducted on a spatialized reverberant version of WSJ0-2mix dataset. Experimental results show that the proposed method outperforms MDC baseline and even better than the oracle ideal binary mask (IBM). |
收录类别 | EI |
七大方向——子方向分类 | 语音识别与合成 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/44388 |
专题 | 多模态人工智能系统全国重点实验室_智能交互 |
通讯作者 | Tao, Jianhua |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 2.CAS Center for Excellence in Brain Science and Intelligence Technology 3.School of Artificial Intelligence, University of Chinese Academy of Sciences |
第一作者单位 | 模式识别国家重点实验室 |
通讯作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Fan, Cunhang,Tao, Jianhua,Liu, Bin,et al. Gated Recurrent Fusion of Spatial and Spectral Features for Multi-channel Speech Separation with Deep Embedding Representations[C],2020. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
interspeech2020-sepa(260KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论