Metraipolis › archive › dc_207af4ce4f7d

Acoustic Progress Propagation for Long-Horizon Speculative Decoding in ASR — arXiv:2609.33245

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33245
License (as archived)
CC0-1.0
Archived
2026-10-07T21:54:00.517Z
SHA-256 of the archived text
74b5228ba2e45bcb2d37a1b6d90dff18d2488ce6440446dc0b5bf5a59005ac77
Stable URL of this page
https://metraipolis.com/documento/dc_207af4ce4f7d

How to cite

Acoustic Progress Propagation for Long-Horizon Speculative Decoding in ASR — arXiv:2609.33245 — arXiv (en), https://arxiv.org/abs/2609.33245 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_207af4ce4f7d (Metraipolis archive dc_207af4ce4f7d, sha256 74b5228ba2e45bcb2d37a1b6d90dff18d2488ce6440446dc0b5bf5a59005ac77)
BibTeX
@misc{dc207af4ce4f7d,
  title = {Acoustic Progress Propagation for Long-Horizon Speculative Decoding in ASR — arXiv:2609.33245},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33245},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_207af4ce4f7d; sha256 74b5228ba2e45bcb2d37a1b6d90dff18d2488ce6440446dc0b5bf5a59005ac77},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33245. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Speculative decoding accelerates autoregressive automatic speech recognition (ASR), but the acceptance length of alignment-aware drafters can saturate as the draft horizon increases. We propose a progress-aware speculative drafter that recurrently propagates an acoustic progress state across draft steps and feeds it back into audio cross-attention to guide token generation. We jointly train the drafter and progress predictor over variable draft horizons. On five ASR test sets, our method achieves lossless, macro-averaged end-to-end speedups of 1.657x and 1.227x over target-only autoregressive decoding with Qwen3-ASR-0.6B and Qwen3-ASR-1.7B, respectively. Relative to AnchorDraft, our method improves the macro-averaged speedup by 34.3% and 9.0%, respectively. Horizon sweeps show continued growth in acceptance length beyond the baselines' saturation. Code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL).
Categorie arXiv: Audio and Speech Processing (eess.AS) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.33245 — 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 2026-10-07T21:27:08.755Z 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).