Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.20042 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918400064749568 |
|---|---|
| author | Chen, Jianan Gao, Xiaoxue Kawahara, Tatsuya Chen, Nancy F. |
| author_facet | Chen, Jianan Gao, Xiaoxue Kawahara, Tatsuya Chen, Nancy F. |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_20042 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LoASR-Bench: Evaluating Large Speech Language Models on Low-Resource Automatic Speech Recognition Across Language Families Chen, Jianan Gao, Xiaoxue Kawahara, Tatsuya Chen, Nancy F. Computation and Language Artificial Intelligence 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. |
| title | LoASR-Bench: Evaluating Large Speech Language Models on Low-Resource Automatic Speech Recognition Across Language Families |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2603.20042 |