用大模型模拟人听觉纠错，提升语音识别准确率。
Listening, Imagining & Refining: A Heuristic Optimized ASR Correction Framework with LLMs
- 模拟人耳听觉过程，分听、想、改三步迭代修正语音转录。
- 在中英文数据上使字符错误率降低最多1.5个百分点。
- 结合有限状态机防止陷入错误修正陷阱，适合高精度场景。
自动语音识别（ASR）系统仍存在影响下游应用的错误。本文提出LIR-ASR，一种受人类听觉感知启发的基于大模型的启发式优化迭代纠错框架。该框架采用“听-想-改”策略，生成语音变体并在上下文中进行优化。引入基于有限状态机（FSM）的启发式优化，避免修正过程陷入局部最优；规则约束确保语义一致性。在中英文ASR输出上的实验表明，相比基线方法，LIR-ASR平均将字符错误率（CER）和词错误率（WER）降低达1.5个百分点，显著提升了转录准确性。
原文摘要 · Abstract (English)
Automatic Speech Recognition (ASR) systems remain prone to errors that affect downstream applications. In this paper, we propose LIR-ASR, a heuristic optimized iterative correction framework using LLMs, inspired by human auditory perception. LIR-ASR applies a "Listening-Imagining-Refining" strategy, generating phonetic variants and refining them in context. A heuristic optimization with finite state machine (FSM) is introduced to prevent the correction process from being trapped in local optima and rule-based constraints help maintain semantic fidelity. Experiments on both English and Chinese ASR outputs show that LIR-ASR achieves average reductions in CER/WER of up to 1.5 percentage points compared to baselines, demonstrating substantial accuracy gains in transcription.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。