arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2105.03095cs.CL
收藏

语篇翻译中共享语义空间的学习

Learning Shared Semantic Space for Speech-to-Text Translation

ByteDance AI Lab(字节跳动人工智能实验室);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Chi Han, Mingxuan Wang, Heng Ji, Lei Li

摘要:

端到端语音翻译(ST)具有众多的潜在应用和巨大的影响,长期以来一直被视为一项独立的任务,未能充分汲取其兄弟-文本机器翻译(MT)的快速发展的力量。由于文本和音频输入的表示方式不同,模态差距使得MT数据及其端到端模型与ST数据不兼容。在观察这个障碍,我们建议弥合这一代表性的差距与嵌合体。通过将音频和文本特征投影到一个共同的语义表示中,Chimera统一了MT和ST任务,并将ST基准测试(Must-C和增强Libripeech)的性能提升到了一个新的最先进水平。具体来说,Chimera在Must-C EN-DE上获得了27.1 BLEU,将SOTA提高了+1.9 BLEU。进一步的实验分析表明,共享的语义空间确实传达了这两个任务之间的共同知识,从而为跨模态增加训练资源铺平了新的道路。代码、数据和资源可在www.example.com上获得。

英文摘要:

Having numerous potential applications and great impact, end-to-end speech translation (ST) has long been treated as an independent task, failing to fully draw strength from the rapid advances of its sibling - text machine translation (MT). With text and audio inputs represented differently, the modality gap has rendered MT data and its end-to-end models incompatible with their ST counterparts. In observation of this obstacle, we propose to bridge this representation gap with Chimera. By projecting audio and text features to a common semantic representation, Chimera unifies MT and ST tasks and boosts the performance on ST benchmarks, MuST-C and Augmented Librispeech, to a new state-of-the-art. Specifically, Chimera obtains 27.1 BLEU on MuST-C EN-DE, improving the SOTA by a +1.9 BLEU margin. Further experimental analyses demonstrate that the shared semantic space indeed conveys common knowledge between these two tasks and thus paves a new way for augmenting training resources across modalities. Code, data, and resources are available at https://github.com/Glaciohound/Chimera-ST.

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。

↑