PaperScope
LIVE · 2026-09-29 05:40 UTC

Acoustic Progress Propagation for Long-Horizon Speculative Decoding in ASR

Yuanyuan Jia, Qianqian Yang

Latestcs.CLcs.LGcs.AIcs.CV
arXiv ID
2609.33245 v1
Category
Submitted
2026-09-27

Abstract

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 https://github.com/yuanyuanjia71-spec/ProgDraft.

Comment: 5 pages, 2 figures, 2 tables

arXiv abs page · PDF · same-day batch