Preprint
Open Universal Arabic ASR Leaderboard
Yingzhi Wang et al., arXiv, 2024
Read the originalRead on October 17, 2026
- What it is
- A 2024 arXiv paper that introduces a continuous benchmark for open-source Arabic speech models tested on several multi-dialect datasets. It is the main public leaderboard for comparing Arabic models across dialects.
- How much weight it carries
- Preprint describing a leaderboard that is updated over time. Figures in the paper are a snapshot, and the live leaderboard is the place to check current results.
Facts taken from this source
- The Open Universal Arabic ASR Leaderboard is a continuous benchmark of open-source Arabic speech models across several multi-dialect datasets.
“a continuous benchmark project for open-source general Arabic ASR models across various multi-dialect datasets” (Abstract)