arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2204.02804cs.SDcs.LGeess.AS
收藏

联邦自我监督语音表示:我们还在吗?

Federated Self-supervised Speech Representations: Are We There Yet?

University of Cambridge(剑桥大学);Samsung AI(三星人工智能);Avignon University(阿维尼翁大学)

Yan Gao, Javier Fernandez-Marques, Titouan Parcollet, Abhinav Mehrotra, Nicholas D. Lane

摘要:

支持麦克风的设备无处不在,导致大量未标记的音频数据在边缘产生。将自监督学习(SSL)和联邦学习(FL)集成到一个连贯的系统中,可以潜在地提供数据隐私保证,同时还可以提高语音表示的质量和鲁棒性。在本文中,我们提供了一个第一的系统研究的可行性和复杂性训练语音SSL模型下FL场景的算法,硬件和系统限制的角度。尽管它们的组合具有很高的潜力,但我们发现现有的系统限制和算法行为使得SSL和FL系统在今天几乎不可能构建。然而,至关重要的是,我们的研究结果表明,具体的性能瓶颈和研究机会,将使这种情况得到扭转。虽然我们的分析表明,考虑到硬件的现有趋势,混合SSL和FL语音系统要到2027年才能实现。我们相信,这项研究可以作为一个路线图,加快工作,实现这一里程碑早得多。

英文摘要:

The ubiquity of microphone-enabled devices has lead to large amounts of unlabelled audio data being produced at the edge. The integration of self-supervised learning (SSL) and federated learning (FL) into one coherent system can potentially offer data privacy guarantees while also advancing the quality and robustness of speech representations. In this paper, we provide a first-of-its-kind systematic study of the feasibility and complexities for training speech SSL models under FL scenarios from the perspective of algorithms, hardware, and systems limits. Despite the high potential of their combination, we find existing system constraints and algorithmic behaviour make SSL and FL systems nearly impossible to build today. Yet critically, our results indicate specific performance bottlenecks and research opportunities that would allow this situation to be reversed. While our analysis suggests that, given existing trends in hardware, hybrid SSL and FL speech systems will not be viable until 2027. We believe this study can act as a roadmap to accelerate work towards reaching this milestone much earlier.

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。

↑