Knowledge Commons of Institute of Automation,CAS
Audio-Visual Speech Separation with Visual Features Enhanced by Adversarial Training | |
Zhang Peng1,2; Xu Jiaming1,2; Shi Jing1; Hao Yunzhe1; Qin Lei4; Xu Bo1,2,3 | |
2021 | |
会议名称 | the 33th International Joint Conference on Neural Networks |
会议录名称 | 0 |
卷号 | 0 |
期号 | 0 |
页码 | 0 |
会议日期 | 2021-7-18 |
会议地点 | 线上会议 |
会议录编者/会议主办者 | INNS-International Neural Network Society ; IEEE-Computational Intelligence Society |
出版地 | 0 |
出版者 | 0 |
产权排序 | 1 |
摘要 | Audio-visual speech separation (AVSS) refers to separating individual voice from an audio mixture of multiple simultaneous talkers by conditioning on visual features. For the AVSS task, visual features play an important role, based on which we manage to extract more effective visual features to improve the performance. In this paper, we propose a novel AVSS model that uses speech-related visual features for isolating the target speaker. Specifically, the method of extracting speechrelated visual features has two steps. Firstly, we extract the visual features that contain speech-related information by learning joint audio-visual representation. Secondly, we use the adversarial training method to enhance speech-related information in visual features further. We adopt the time-domain approach and build audio-visual speech separation networks with temporal convolutional neural networks block. Experiments on four audio-visual datasets, including GRID, TCD-TIMIT, AVSpeech, and LRS2, show that our model significantly outperforms previous state-ofthe-art AVSS models. We also demonstrate that our model can achieve excellent speech separation performance in noisy realworld scenarios. Moreover, in order to alleviate the performance degradation of AVSS models caused by the missing of some video frames, we propose a training strategy, which makes our model robust when video frames are partially missing. The demo, code, and supplementary materials can be available at https://github.com/aispeech-lab/advr-avss |
关键词 | audio-visual speech separation robust adversarial training method time-domain approach |
学科门类 | 工学::控制科学与工程 |
DOI | 0 |
URL | 查看原文 |
收录类别 | EI |
语种 | 英语 |
七大方向——子方向分类 | 语音识别与合成 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/44910 |
专题 | 复杂系统认知与决策实验室_听觉模型与认知计算 中国科学院自动化研究所 |
通讯作者 | Xu Jiaming; Xu Bo |
作者单位 | 1.Institute of Automation, Chinese Academy of Science 2.School of Artificial Intelligence, University of Chinese Academy of Science 3.Center for Excellence in Brain Science and Intelligence Technology 4.Huawei Consumer Business Group |
推荐引用方式 GB/T 7714 | Zhang Peng,Xu Jiaming,Shi Jing,et al. Audio-Visual Speech Separation with Visual Features Enhanced by Adversarial Training[C]//INNS-International Neural Network Society, IEEE-Computational Intelligence Society. 0:0,2021:0. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
2021070955.pdf(1900KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
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