MauBERT: Universal Phonetic Inductive Biases for Few-Shot Acoustic Units Discovery
This provides a more robust approach for cross-lingual speech processing, though it builds incrementally on existing HuBERT methodology.
The paper tackles the problem of learning language-independent phonetic representations by extending HuBERT with articulatory feature supervision across 55 languages, resulting in models that outperform state-of-the-art multilingual self-supervised models in ABX discriminability tests and adapt to unseen languages with only 10 hours of fine-tuning.
This paper introduces MauBERT, a multilingual extension of HuBERT that leverages articulatory features for robust cross-lingual phonetic representation learning. We continue HuBERT pre-training with supervision based on a phonetic-to-articulatory feature mapping in 55 languages. Our models learn from multilingual data to predict articulatory features or phones, resulting in language-independent representations that capture multilingual phonetic properties. Through comprehensive ABX discriminability testing, we show MauBERT models produce more context-invariant representations than state-of-the-art multilingual self-supervised learning models. Additionally, the models effectively adapt to unseen languages and casual speech with minimal self-supervised fine-tuning (10 hours of speech). This establishes an effective approach for instilling linguistic inductive biases in self-supervised speech models.