Metraipolis › archive › dc_633fdec76bc3

Language Discrimination Improves Linguistic Learning in Multilingual Speech Models — arXiv:2609.33345

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.33345
License (as archived): CC0-1.0
Archived: 2026-10-07T21:52:57.219Z
SHA-256 of the archived text: 84598e5fe11b2b4075f80a3e35163c4ac5873dc2aa6c89a75e4fa9fd27af13fa
Stable URL of this page: https://metraipolis.com/documento/dc_633fdec76bc3

How to cite

Language Discrimination Improves Linguistic Learning in Multilingual Speech Models — arXiv:2609.33345 — arXiv (en), https://arxiv.org/abs/2609.33345 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_633fdec76bc3 (Metraipolis archive dc_633fdec76bc3, sha256 84598e5fe11b2b4075f80a3e35163c4ac5873dc2aa6c89a75e4fa9fd27af13fa)

BibTeX

@misc{dc633fdec76bc3,
  title = {Language Discrimination Improves Linguistic Learning in Multilingual Speech Models — arXiv:2609.33345},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33345},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_633fdec76bc3; sha256 84598e5fe11b2b4075f80a3e35163c4ac5873dc2aa6c89a75e4fa9fd27af13fa},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33345. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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, lower is better) decreases from 11.6% in the bilingual baseline to 10.4% (monolingual: 10.8%), while lexical performance (sWUGGY, higher is better) increases from 52.1% to 56.7% (monolingual: 58.5%) and prosodic performance (ProsAudit, lexical subtask, higher is better) from 68.9% to 72.9% (monolingual: 72.6%). Across HuBERT training stages, the strongest gains on most linguistic measures occur when language discrimination is introduced in the first iteration, whereas later or repeated interventions yield smaller improvements and are accompanied by increased language-wise segregation. These results support a causal role for language discrimination in reducing the additional cost of multilingual learning.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Sound (cs.SD); Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2609.33345 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T21:27:08.511Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).