When Multiple Scripts Matter: leaderboard
Automatic speech recognition (ASR) in non-English clinical settings is challenged by multiscript variability, where the same term may appear in multiple valid orthographic forms. Conventional string-matching evaluation metrics often underestimate ASR performance by treating orthographic variants as errors. To address this issue, we introduce MultiClin, a clinical ASR benchmark designed to evaluate
Metric: CER (original, original, original) (self-reported). Source: benchmarklist.com. Saturation forecast: Around October 2026. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 24.23 | #152 |
| 2 | Gemini 2.5 Flash | 24.87 | #245 |
| 3 | Qwen 3 1.7B | 29.73 | #1196 |
| 4 | Qwen 3 0.6B | 59.09 | #1459 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=when-multiple-scripts-matter · How It Works · Data refreshed daily, snapshot 2026-10-20.