DeRAGEC: Denoising Named Entity Candidates with Synthetic Rationale for ASR Error Correction
Preprint
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/deragecKeywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Retrieval Augmented Correction of Named Entity Speech Recognition Errors2025
- Contextualization of ASR with LLM using phonetic retrieval-based augmentation2025
- Failing Forward: Improving Generative Error Correction for ASR with Synthetic Data and Retrieval Augmentation2025
- Investigating ASR Error Correction with Large Language Model and Multilingual 1-best Hypotheses2024
- DANCER: Entity Description Augmented Named Entity Corrector for Automatic Speech Recognition2024
- HyPoradise: An Open Baseline for Generative Speech Recognition with Large Language Models2023
- Large Language Models with Controllable Working Memory2023