Research

Paper

AI LLM March 20, 2026

LoASR-Bench: Evaluating Large Speech Language Models on Low-Resource Automatic Speech Recognition Across Language Families

Authors

Jianan Chen, Xiaoxue Gao, Tatsuya Kawahara, Nancy F. Chen

Abstract

Large language models (LLMs) have driven substantial advances in speech language models (SpeechLMs), yielding strong performance in automatic speech recognition (ASR) under high-resource conditions. However, existing benchmarks predominantly focus on high-resource languages, leaving the ASR behavior of SpeechLMs in low-resource languages insufficiently understood. This gap is critical, as practical ASR systems must reliably support low-resource languages and generalize across diverse language families, and it directly hinders the deployment of SpeechLM-based ASR in real-world multilingual scenarios. As a result, it is essential to evaluate SpeechLMs on low-resource languages to ensure their generalizability across different language families. To address this problem, we propose \textbf{LoASR-Bench}, a comprehensive benchmark designed to evaluate \textbf{lo}w-resource \textbf{a}utomatic \textbf{s}peech \textbf{r}ecognition (\textbf{ASR}) of the latest SpeechLMs across diverse language families. LoASR-Bench comprises 25 languages from 9 language families, featuring both Latin and non-Latin scripts, enabling cross-linguistic and cross-script assessment of ASR performance of current SpeechLMs. Experimental results highlight the limitations of the latest SpeechLMs in handling real-world low-resource languages.

Metadata

arXiv ID: 2603.20042
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-20
Fetched: 2026-03-23 16:54

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.20042v1</id>\n    <title>LoASR-Bench: Evaluating Large Speech Language Models on Low-Resource Automatic Speech Recognition Across Language Families</title>\n    <updated>2026-03-20T15:26:34Z</updated>\n    <link href='https://arxiv.org/abs/2603.20042v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.20042v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Large language models (LLMs) have driven substantial advances in speech language models (SpeechLMs), yielding strong performance in automatic speech recognition (ASR) under high-resource conditions. However, existing benchmarks predominantly focus on high-resource languages, leaving the ASR behavior of SpeechLMs in low-resource languages insufficiently understood. This gap is critical, as practical ASR systems must reliably support low-resource languages and generalize across diverse language families, and it directly hinders the deployment of SpeechLM-based ASR in real-world multilingual scenarios. As a result, it is essential to evaluate SpeechLMs on low-resource languages to ensure their generalizability across different language families. To address this problem, we propose \\textbf{LoASR-Bench}, a comprehensive benchmark designed to evaluate \\textbf{lo}w-resource \\textbf{a}utomatic \\textbf{s}peech \\textbf{r}ecognition (\\textbf{ASR}) of the latest SpeechLMs across diverse language families. LoASR-Bench comprises 25 languages from 9 language families, featuring both Latin and non-Latin scripts, enabling cross-linguistic and cross-script assessment of ASR performance of current SpeechLMs. Experimental results highlight the limitations of the latest SpeechLMs in handling real-world low-resource languages.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-20T15:26:34Z</published>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Jianan Chen</name>\n    </author>\n    <author>\n      <name>Xiaoxue Gao</name>\n    </author>\n    <author>\n      <name>Tatsuya Kawahara</name>\n    </author>\n    <author>\n      <name>Nancy F. Chen</name>\n    </author>\n  </entry>"
}