--- format: "aidr-story-markdown/v1" id: "bd7fa85e5bd0311f32d40a6179bd30f377e2b8a9ef9e065c98055aeabfcd1d5c" canonical_url: "https://aidr.today/bd7fa85e?lang=en" title: "Language Discrimination Improves Linguistic Learning in Multilingual Speech Models" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-10-02T00:00:00.000Z" category: "Research" topics: ["meta","llm"] source_urls: ["https://machinelearning.apple.com/research/language-discrimination-multilingual-learning"] summary: "Language Discrimination Improves Linguistic Learning in Multilingual Speech Models Authors Maureen de Seyssel, Jie Chi*, Zakaria Aldeneh* Multilingual self-supervised speech models can benefit from sharing information across languages, but under a matched total pretraining data budget they still fall short of monolingual models. We show that strengthening the model’s ability to discriminate languages during pretraining reduces and, on some measures, closes this multilingual gap on continuous phonetic and higher-level linguistic measures, while preserving substantial cross-language sharing. Using a controlled English/French HuBERT setting, we test two interventions which strengthen language discrimination: an auxiliary language classifier and per-language k-means targets. Across interventions, continuous-feature phone discrimination error (phone-ABX,↓) decreases from 11.6% in the bilingual baseline to 10.4% (monolingual: 10.8%), while lexical performance (sWUGGY,↑) increases from 52.1% to 56.7% (monolingual: 58.5%) and prosodic performance (ProsAudit, lexical subtask,↑) from 68.9% to 72.9% (monolingual: 72.6%). Across HuBERT training stages, the strongest gains on most linguistic m…" --- # Language Discrimination Improves Linguistic Learning in Multilingual Speech Models > [Open the canonical story]() **Published:** 2026-10-02T00:00:00.000Z **Category:** Research **Topics:** meta, llm ## Summary Language Discrimination Improves Linguistic Learning in Multilingual Speech Models Authors Maureen de Seyssel, Jie Chi\*, Zakaria Aldeneh\* Multilingual self\-supervised speech models can benefit from sharing information across languages, but under a matched total pretraining data budget they still fall short of monolingual models\. We show that strengthening the model’s ability to discriminate languages during pretraining reduces and, on some measures, closes this multilingual gap on continuous phonetic and higher\-level linguistic measures, while preserving substantial cross\-language sharing\. Using a controlled English/French HuBERT setting, we test two interventions which strengthen language discrimination: an auxiliary language classifier and per\-language k\-means targets\. Across interventions, continuous\-feature phone discrimination error \(phone\-ABX,↓\) decreases from 11\.6% in the bilingual baseline to 10\.4% \(monolingual: 10\.8%\), while lexical performance \(sWUGGY,↑\) increases from 52\.1% to 56\.7% \(monolingual: 58\.5%\) and prosodic performance \(ProsAudit, lexical subtask,↑\) from 68\.9% to 72\.9% \(monolingual: 72\.6%\)\. Across HuBERT training stages, the strongest gains on most linguistic m… ## Sources - [Story source]()