Metraipolis › archive › dc_ecfca9263424

Mitigating Accent-Language Confusion in Self-Supervised Speech Representations for Language Identification — arXiv:2610.09486

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09486
License (as archived)
CC0-1.0
Archived
2026-10-08T11:26:33.767Z
SHA-256 of the archived text
49f30c03f498bee7ab5f6506ccf00c24c8419fd8427da606deb2472551a991c2
Stable URL of this page
https://metraipolis.com/documento/dc_ecfca9263424

How to cite

Mitigating Accent-Language Confusion in Self-Supervised Speech Representations for Language Identification — arXiv:2610.09486 — arXiv (en), https://arxiv.org/abs/2610.09486 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ecfca9263424 (Metraipolis archive dc_ecfca9263424, sha256 49f30c03f498bee7ab5f6506ccf00c24c8419fd8427da606deb2472551a991c2)
BibTeX
@misc{dcecfca9263424,
  title = {Mitigating Accent-Language Confusion in Self-Supervised Speech Representations for Language Identification — arXiv:2610.09486},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09486},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ecfca9263424; sha256 49f30c03f498bee7ab5f6506ccf00c24c8419fd8427da606deb2472551a991c2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09486. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Spoken language identification (LID) aims to recognize the target language regardless of accent. In practice, however, LID models fine-tuned from self-supervised speech representations frequently confuse accents with languages, misclassifying non-native (L2) speech as the speaker's first language (L1). We show that non-native speech representations lie between native target-language and native L1 poles, causing systematic misclassification. To address this, we introduce a geometric projection that estimates an L1-bias direction solely from native speech and removes it before the frozen LID head. Across five MMS-LID models and non-native corpora, this projection substantially improves target language identification for L2-accented speech while preserving predictions for native speech. These results show that accent-induced L1 bias can be corrected directly within the representation space without L2 training data or model adaptation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Audio and Speech Processing (eess.AS) ; Computation and Language (cs.CL); Sound (cs.SD).
Pagina del lavoro: https://arxiv.org/abs/2610.09486 — 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 2026-10-08T10:24:06.379Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).