pith. sign in

arxiv: 2303.09455 · v1 · pith:O5BANN5Lnew · submitted 2023-03-14 · 💻 cs.CL · cs.CV· cs.LG· cs.SD· eess.AS

Learning Cross-lingual Visual Speech Representations

classification 💻 cs.CL cs.CVcs.LGcs.SDeess.AS
keywords learningcross-lingualdataspeechvisualaudio-visualbetterlanguages
0
0 comments X
read the original abstract

Cross-lingual self-supervised learning has been a growing research topic in the last few years. However, current works only explored the use of audio signals to create representations. In this work, we study cross-lingual self-supervised visual representation learning. We use the recently-proposed Raw Audio-Visual Speech Encoders (RAVEn) framework to pre-train an audio-visual model with unlabelled multilingual data, and then fine-tune the visual model on labelled transcriptions. Our experiments show that: (1) multi-lingual models with more data outperform monolingual ones, but, when keeping the amount of data fixed, monolingual models tend to reach better performance; (2) multi-lingual outperforms English-only pre-training; (3) using languages which are more similar yields better results; and (4) fine-tuning on unseen languages is competitive to using the target language in the pre-training set. We hope our study inspires future research on non-English-only speech representation learning.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.