Metraipolis › archive › dc_7185771fcba5
Merging the Knowledge of LLMs for Automatic Speech Recognition — arXiv:2609.15743
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.15743
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T23:16:15.736Z
- SHA-256 of the archived text
793bff391d5cce0db5f5cd9e719b0dc4195cfcd573f94f8989a6370821b591eb- Stable URL of this page
https://metraipolis.com/documento/dc_7185771fcba5
How to cite
Merging the Knowledge of LLMs for Automatic Speech Recognition — arXiv:2609.15743 — arXiv (en), https://arxiv.org/abs/2609.15743 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7185771fcba5 (Metraipolis archive dc_7185771fcba5, sha256 793bff391d5cce0db5f5cd9e719b0dc4195cfcd573f94f8989a6370821b591eb)
BibTeX
@misc{dc7185771fcba5,
title = {Merging the Knowledge of LLMs for Automatic Speech Recognition — arXiv:2609.15743},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.15743},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7185771fcba5; sha256 793bff391d5cce0db5f5cd9e719b0dc4195cfcd573f94f8989a6370821b591eb},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15743. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.15743 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T22:51:09.955Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).