DeRAGEC: Denoising Named Entity Candidates with Synthetic Rationale for ASR Error Correction

2025/06/09 by Sungjin Im, Im, Solee, Wonjun Lee +9 · 1 citation
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Speech Recognition and Synthesis #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2506.07510

openalex publication_date 2025/06/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present DeRAGEC, a method for improving Named Entity (NE) correction in Automatic Speech Recognition (ASR) systems. By extending the Retrieval-Augmented Generative Error Correction (RAGEC) framework, DeRAGEC employs synthetic denoising rationales to filter out noisy NE candidates before correction. By leveraging phonetic similarity and augmented definitions, it refines noisy retrieved NEs using in-context learning, requiring no additional training. Experimental results on CommonVoice and STOP datasets show significant improvements in Word Error Rate (WER) and NE hit ratio, outperforming baseline ASR and RAGEC methods. Specifically, we achieved a 28% relative reduction in WER compared to ASR without postprocessing. Our source code is publicly available at: https://github.com/solee0022/deragec

Citations

• Failing Forward: Improving Generative Error Correction for ASR with Synthetic Data and Retrieval Augmentation
• Contextualization of ASR with LLM using phonetic retrieval-based augmentation
• Retrieval Augmented Correction of Named Entity Speech Recognition Errors
• The Llama 3 Herd of Models
• Making Retrieval-Augmented Language Models Robust to Irrelevant Context
• Bias and Fairness in Large Language Models: A Survey
• Bias and Fairness in Large Language Models: A Survey
• Can Generative Large Language Models Perform ASR Error Correction?
• Robust Speech Recognition via Large-Scale Weak Supervision
• Large Language Models with Controllable Working Memory
• Atlas: Few-shot Learning with Retrieval Augmented Language Models
• STOP: A dataset for Spoken Task Oriented Semantic Parsing
• Space-Efficient Representation of Entity-centric Query Language Models
• Language Models are Few-Shot Learners
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• REALM: Retrieval-Augmented Language Model Pre-Training
• Common Voice: A Massively-Multilingual Speech Corpus
• InstructRAG: Instructing Retrieval-Augmented Generation via Self-Synthesized Rationales

Cited by

• A Survey on Non-Intrusive ASR Refinement: From Output-Level Correction to Full-Model Distillation

Related

• Failing Forward: Improving Generative Error Correction for ASR with Synthetic Data and Retrieval Augmentation
• Generative Annotation for ASR Named Entity Correction
• DANCER: Entity Description Augmented Named Entity Corrector for Automatic Speech Recognition
• Retrieval Augmented Correction of Named Entity Speech Recognition Errors
• Improving Speech Recognition of Named Entities in Classroom Speech with LLM Revision and Phonetic-Semantic Context