---
title: "REDDIT：通过重播式分布编辑纠正ASR中的模型生成时间戳漂移，而不遗忘已学知识"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.05364v1"
permalink: "https://feed.astralor.com/items/itm_6fa11e7e91fc8f45bb84bd24"
published: "2026-07-06T17:40:54.000Z"
score: 80
channel: "ai"
tags: ["ASR","OpenAI","Whisper-tiny","后训练框架","时间戳漂移","灾难性遗忘"]
---

# REDDIT：通过重播式分布编辑纠正ASR中的模型生成时间戳漂移，而不遗忘已学知识

## AI 点评

> 这篇论文对ASR从业者非常值得一读，特别是那些处理长音频或实时字幕任务的团队。它解决了实际部署中常见但常被忽略的时间戳漂移问题，而且方法轻量、无需昂贵标注。如果你在用Whisper或其他端到端ASR模型，REDDIT的思路可以直接借鉴来提升时间对齐精度。

## 速览

现代自回归ASR系统能将时间戳作为解码令牌输出，实现无需帧级对齐的时间戳转录。然而，这些生成的时间戳在长无声片段中会出现漂移：转录可能仍合理，但解码时间轴偏离音频。本文研究了这一问题，并提出了REDDIT（基于重播的分布编辑），一个轻量级两阶段后训练框架，通过编辑时间戳目标并避免灾难性遗忘，在仅更新1.6%参数且使用34.9小时音频的情况下，将长间隙mIoU从38.7%提升至95.0%，混合同外域AAS从2752ms降至223ms，同时保持CV-en MER在41.3%水平。

## 关键点

- 自回归ASR系统在长无声段中会产生时间戳漂移，导致转录合理但时间轴偏移。
- 简单的时间戳纠正微调会严重损害非目标ASR行为，引发灾难性遗忘。
- REDDIT通过重播解码上下文编辑时间戳目标，并匹配冻结基分布，无需人工标注即可构建校正监督。
- 在Whisper-tiny上仅更新1.6%参数，将长间隙mIoU提升至95%，几乎消除漂移。

## 全文

Modern autoregressive ASR systems can emit timestamps as decoded tokens, enabling timestamped transcription without frame-level aligners or inference-time post-processing. We show that these generated timestamps can drift across long non-speech spans: the transcript may remain plausible, but the decoded time axis drifts away from the audio. We study this non-speech-induced timestamp drift with self-built gap and long-gap benchmarks across 15 evaluated timestamp-producing ASR and audio-language systems. Naive timestamp-corrected fine-tuning improves alignment but can severely degrade non-target ASR behavior, exposing a forgetting problem. We propose REDDIT(REplay-based Distribution eDITing), a lightweight two-stage post-training framework that corrects timestamps while avoiding this catastrophic forgetting: it first edits timestamp targets under the model's own replayed decoder context while matching the frozen base distribution on non-timestamp tokens, then applies a short edited-prefix refinement stage. In this framework, we construct correction supervision without human transcripts or human timestamp annotations by combining VAD-trimmed speech spans with inserted non-speech gaps and known concatenation offsets. On Whisper-tiny, 34.9 hours of targeted correction audio used and only 1.6% of model parameters updated, raising long-gap mIoU from 38.7% to 95.0% and reducing mixed-gap out-of-domain AAS from 2752 ms to 223 ms while preserving CV-en MER at 41.3% (versus 524.2% for ordinary SFT decoder tuning).

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_6fa11e7e91fc8f45bb84bd24