Cross-Lingual Speech Model Performance Gains from Expanded Flemish Dutch Pre-training Data

Multilingual BERT (mBERT) However, these evaluations have focused on cross-lingual transfer with highresource languages, covering only a third of the languages covered by mBERT. We explore how mBERT performs on a much wider set of languages, focusing on the quality of representation for low-resource languages, measured by within-language performance. We consider three tasks: Named Entity Recognition (99 languages), Part-of-speech Tagging, and Dependency Parsing (54 languages each). mBERT does better than or comparable to baselines on high resource languages but does much worse for low resource

Research goal: To what extent does increasing the amount of Flemish Dutch pre-training data improve the cross-lingual transfer performance of speech models, as evaluated by WER on the LibriSpeech and VoxForge corpora for both high- and low-resource languages?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.7/10.