Unsupervised ASR via Cross-Lingual Pseudo-Labeling

arXiv:2305.13330v3 Announce Type: replace-cross
Abstract: Recent work has shown that it is possible to train an $\textit{unsupervised}$ automatic speech recognition (ASR) system using only unpaired audio and text. Existing unsupervised ASR methods assume that no labeled data can be used for training. We argue that even if one does not have any labeled audio for a given language, there is $\textit{always}$ labeled data available for other languages. We show that it is possible to use character-level acoustic models (AMs) from …

abstract arxiv asr audio automatic speech recognition cross-lingual cs.cl cs.lg cs.sd data eess.as labeling recognition speech speech recognition text train training type unsupervised via work