Research

Paper

AI LLM March 04, 2026

Robust LLM-based Audio-Visual Speech Recognition with Sparse Modality Alignment and Visual Unit-Guided Refinement

Authors

Fei Su, Cancan Li, Juan Liu, Wei Ju, Hongbin Suo, Ming Li

Abstract

Audio-Visual Speech Recognition (AVSR) integrates acoustic and visual information to enhance robustness in adverse acoustic conditions. Recent advances in Large Language Models (LLMs) have yielded competitive automatic speech recognition performance and shown effectiveness for AVSR. However, prior approaches project audio and visual features independently or apply shallow fusion, limiting cross-modal alignment and complementary exchange while increasing the LLM's computational load. To address this, we propose AVUR-LLM, an LLM-based Audio-Visual Speech Recognition via Sparse Modality Alignment and Visual Unit-Guided Refinement. Experiments on LRS3 demonstrate state-of-the-art results for AVSR. Under additive-noise conditions at 0 dB SNR, it achieves 37% relative improvement over the baseline system.

Metadata

arXiv ID: 2603.03811
Provider: ARXIV
Primary Category: cs.SD
Published: 2026-03-04
Fetched: 2026-03-05 06:06

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.03811v1</id>\n    <title>Robust LLM-based Audio-Visual Speech Recognition with Sparse Modality Alignment and Visual Unit-Guided Refinement</title>\n    <updated>2026-03-04T07:47:30Z</updated>\n    <link href='https://arxiv.org/abs/2603.03811v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.03811v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Audio-Visual Speech Recognition (AVSR) integrates acoustic and visual information to enhance robustness in adverse acoustic conditions. Recent advances in Large Language Models (LLMs) have yielded competitive automatic speech recognition performance and shown effectiveness for AVSR. However, prior approaches project audio and visual features independently or apply shallow fusion, limiting cross-modal alignment and complementary exchange while increasing the LLM's computational load. To address this, we propose AVUR-LLM, an LLM-based Audio-Visual Speech Recognition via Sparse Modality Alignment and Visual Unit-Guided Refinement. Experiments on LRS3 demonstrate state-of-the-art results for AVSR. Under additive-noise conditions at 0 dB SNR, it achieves 37% relative improvement over the baseline system.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SD'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.MM'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='eess.AS'/>\n    <published>2026-03-04T07:47:30Z</published>\n    <arxiv:comment>submitted to Interspeech 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.SD'/>\n    <author>\n      <name>Fei Su</name>\n    </author>\n    <author>\n      <name>Cancan Li</name>\n    </author>\n    <author>\n      <name>Juan Liu</name>\n    </author>\n    <author>\n      <name>Wei Ju</name>\n    </author>\n    <author>\n      <name>Hongbin Suo</name>\n    </author>\n    <author>\n      <name>Ming Li</name>\n    </author>\n  </entry>"
}