Research
Merging the Knowledge of LLMs for Automatic Speech Recognition
arXiv:2609.15743v1 Announce Type: new Abstract: Automatic speech recognition (ASR) systems, trained on paired speech-text data, have been improved by leveraging language models (LMs) trained on text-o
arXiv:2609.15743v1 Announce Type: new Abstract: Automatic speech recognition (ASR) systems, trained on paired speech-text data, have been improved by leveraging language models (LMs) trained on text-only data. LM fusion methods such as shallow fusion and density ratio are well-established methods that incorporate external LMs during ASR decoding. However, they incur additional computational costs due to LM inference, which is particularly problematic for recent larger LMs. In this study, we propose incorporating external LMs via model merging. This method integrates the LMs directly into the parameters of an LLM-based ASR model, requiring no additional computational cost at inference. We formulate domain extension and transfer via arithmetic operations on LoRA parameters. Experimental evaluations were conducted for the domain adaptation of LLM-based ASR trained on CSJ and LibriSpeech. We show that our LM merging consistently improved the ASR performance in the target domains, without degrading inference speed or memory footprint.
Related
- MEUSLI: a Multilingual Projector for LLM-based ASR and Beyond
- Evaluation of Automatic Speech Recognition Using Generative Large Language Models
- Listening or Reading? Evaluating Speech Awareness in Chain-of-Thought Speech-to-Text Translation
- Proactive for Uncertainty: Cause-Aware Error Diagnosis and Interactive Clarification for Spoken Dialogue Systems
Source: arXiv cs.CL | 2026-09-15