Metraipolis › archive › dc_cb1492227d26
Ruby-ASR: Evidence-Preserving Supervision for Joint Orthographic and Lexical-Reading Recognition — arXiv:2609.27289
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.27289
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T09:12:23.065Z
- SHA-256 of the archived text
f051d57d60c80565a228a77eee2895df324ebc84052e12c7bc30c3a135c17f6b- Stable URL of this page
https://metraipolis.com/documento/dc_cb1492227d26
How to cite
Ruby-ASR: Evidence-Preserving Supervision for Joint Orthographic and Lexical-Reading Recognition — arXiv:2609.27289 — arXiv (en), https://arxiv.org/abs/2609.27289 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cb1492227d26 (Metraipolis archive dc_cb1492227d26, sha256 f051d57d60c80565a228a77eee2895df324ebc84052e12c7bc30c3a135c17f6b)
BibTeX
@misc{dccb1492227d26,
title = {Ruby-ASR: Evidence-Preserving Supervision for Joint Orthographic and Lexical-Reading Recognition — arXiv:2609.27289},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.27289},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cb1492227d26; sha256 f051d57d60c80565a228a77eee2895df324ebc84052e12c7bc30c3a135c17f6b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27289. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Conventional Japanese automatic speech recognition (ASR) is supervised by an orthographic transcript, although the same written form can correspond to different lexical readings realized in speech. Such utterances receive an identical target, so their reading distinction is absent from the supervision interface and cannot be recovered reliably by post-hoc text-only grapheme-to-phoneme conversion. We present Ruby-ASR, which refines the conventional target into a span-bound orthographic--lexical-reading sequence. Unlike separate full-sentence orthographic and phonological outputs, the ruby representation locally binds each written span to its realized reading and permits deterministic recovery of both views. We instantiate the target under subtitle-style and verbatim-style transcription conventions using a Qwen3-ASR backbone; a mora-level CTC objective provides auxiliary monotonic reading supervision. The experimental results across five Japanese benchmarks show that refining the recognition target can improve lexical-reading recovery without sacrificing readable orthographic transcription. We release the checkpoints and inference code. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.27289 — 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 2026-10-08T08:13:26.642Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).