What do self-supervised speech models know about Dutch? Analyzing advantages of language-specific pre-training
This work addresses the problem of optimizing speech models for specific languages, which is incremental as it builds on existing self-supervised methods to show language-specific advantages.
The study investigated whether pre-training self-supervised speech models on Dutch improves the encoding of Dutch phonetic and lexical information compared to pre-training on English or multilingual data, finding that language-specific pre-training enhances linguistic feature representation and aligns with better downstream Automatic Speech Recognition performance.
How language-specific are speech representations learned by self-supervised models? Existing work has shown that a range of linguistic features can be successfully decoded from end-to-end models trained only on speech recordings. However, it's less clear to what extent pre-training on specific languages improves language-specific linguistic information. Here we test the encoding of Dutch phonetic and lexical information in internal representations of self-supervised Wav2Vec2 models. Pre-training exclusively on Dutch improves the representation of Dutch linguistic features as compared to pre-training on similar amounts of English or larger amounts of multilingual data. This language-specific advantage is well-detected by trained clustering or classification probes, and partially observable using zero-shot metrics. Furthermore, the language-specific benefit on linguistic feature encoding aligns with downstream performance on Automatic Speech Recognition.