BriefGPT.xyz
May, 2022
利用伪标记数据提高直接语音翻译质量
Leveraging Pseudo-labeled Data to Improve Direct Speech-to-Speech Translation
HTML
PDF
Qianqian Dong, Fengpeng Yue, Tom Ko, Mingxuan Wang, Qibing Bai...
TL;DR
本文介绍了一种基于 Transformer 神经网络的直接语音翻译技术，利用伪标签和外部数据来提升翻译效果，并在 Fisher 英西测试数据上创造了最好的翻译结果
Abstract
Direct
speech-to-speech translation
(S2ST) has drawn more and more attention recently. The task is very challenging due to
data scarcity
and complex speech-to-speech mapping. In this paper, we report our recent a
→