arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.05830eess.AScs.CVcs.LGcs.SDeess.IV
收藏

从推理到生成:端到端完全自监督的语音人脸生成

From Inference to Generation: End-to-end Fully Self-supervised Generation of Human Face from Speech

Seoul National University(首尔大学);Hong Kong University of Science and Technology(香港科技大学)

Hyeong-Seok Choi, Changdae Park, Kyogu Lee

摘要:

这项工作寻求的可能性,从语音生成的人的脸完全基于视听数据,没有任何人类标记的注释。为此,我们提出了一个多模态学习框架,链接的推理阶段和生成阶段。首先,训练推理网络以匹配两种不同模态之间的说话人身份。然后,经过训练的推理网络通过给出关于语音的条件信息与生成网络合作。该方法利用了GANs技术的最新发展,直接从语音波形中生成人脸,使我们的系统完全端到端。我们分析了网络在多大程度上可以自然地解开两个潜在的因素,有助于生成一个人脸图像-一个直接来自语音信号,另一个是不相关的-并探讨网络是否可以学习生成自然的人脸图像分布通过建模这些因素。实验结果表明,该网络不仅能够匹配人脸与语音之间的关系,而且能够根据语音生成高质量的人脸样本。最后,定量测量生成的人脸和相应的语音之间的相关性,以分析两种模态之间的关系。

英文摘要:

This work seeks the possibility of generating the human face from voice solely based on the audio-visual data without any human-labeled annotations. To this end, we propose a multi-modal learning framework that links the inference stage and generation stage. First, the inference networks are trained to match the speaker identity between the two different modalities. Then the trained inference networks cooperate with the generation network by giving conditional information about the voice. The proposed method exploits the recent development of GANs techniques and generates the human face directly from the speech waveform making our system fully end-to-end. We analyze the extent to which the network can naturally disentangle two latent factors that contribute to the generation of a face image - one that comes directly from a speech signal and the other that is not related to it - and explore whether the network can learn to generate natural human face image distribution by modeling these factors. Experimental results show that the proposed network can not only match the relationship between the human face and speech, but can also generate the high-quality human face sample conditioned on its speech. Finally, the correlation between the generated face and the corresponding speech is quantitatively measured to analyze the relationship between the two modalities.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑