BriefGPT.xyz
Oct, 2022
无监督口音领域自适应学习不变表示和风险最小化
Learning Invariant Representation and Risk Minimized for Unsupervised Accent Domain Adaptation
HTML
PDF
Chendong Zhao, Jianzong Wang, Xiaoyang Qu, Haoqian Wang, Jing Xiao
TL;DR
本文探讨通过将语音表征映射到对应的高级语言信息以学习领域不变的语音表征，结果证明，学习到的latents 不仅捕捉到每个音素的发音特征，而且提高了适应能力，在accened测试基准上大幅优于基准模型。
Abstract
unsupervised representation learning
for
speech audios
attained impressive performances for speech recognition tasks, particularly when annotated speech is limited. However, the unsupervised paradigm needs to be
→