# Ruby-ASR paper targets Japanese reading ambiguity with a span-bound transcript format

> An arXiv paper proposes Ruby-ASR, which pairs each written Japanese span with its realized reading instead of a single orthographic transcript.

- **Topic**: Models
- **Published**: 2026-09-24T07:50:17.401Z
- **Canonical URL**: https://highsignal.sh/stories/ruby-asr-paper-targets-japanese-reading-ambiguity-with-a-span-bound-transcript-f-14ae493d

## Why It Matters

Standard Japanese ASR training collapses different spoken readings of the same written form into one target, and the paper argues that gap can't be reliably repaired later by text-only grapheme-to-phoneme conversion.

## Key Findings & Analysis

### What the paper proposes

> **ArXiv CS.CL** describes Ruby-ASR, which refines the conventional Japanese ASR training target from a plain orthographic transcript into a span-bound orthographic–lexical-reading sequence. In the authors' framing, binding each written span locally to its realized reading — rather than emitting separate full-sentence orthographic and phonological outputs — allows deterministic recovery of both views. The paper says it instantiates this target under both subtitle-style and verbatim-style transcription conventions, using a Qwen3-ASR backbone, with a mora-level CTC objective providing auxiliary monotonic reading supervision.

### Reported results and caveats

> According to the **arXiv** abstract, experiments across five Japanese benchmarks show that refining the recognition target can improve lexical-reading recovery without sacrificing readable orthographic transcription. These are the authors' reported results from their own benchmarks; no independent replication or third-party evaluation is included in the supplied evidence. The abstract does not list specific benchmark scores or dataset names. Checkpoints and inference code are stated to be released.

## Primary Sources & Citations

- [Ruby-ASR: Evidence-Preserving Supervision for Joint Orthographic and Lexical-Reading Recognition](https://arxiv.org/abs/2609.27289) — *ArXiv CS.CL (Computation and Language)* (Reporting)

---

[← Back to front page](https://highsignal.sh/) | [Daily Brief](https://highsignal.sh/brief) | [All stories](https://highsignal.sh/latest)