AudioBench - MELD Emotion: leaderboard
AudioBench MELD Emotion slice measuring speech emotion recognition from audio inputs.
Metric: Mean Score (%). Source: github.com. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | SeaLLMs-Audio-7B | 51.11 | #1018 |
| 2 | Qwen-Audio-Chat | 50.73 | #1244 |
| 3 | Phi-4 Multimodal Instruct | 43.52 | #906 |
| 4 | Qwen2-Audio-7B-Instruct | 40.3 | #880 |
| 5 | SALMONN-7B | 30.54 | #1293 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=audiobench-meld-emotion · How It Works · Data refreshed daily, snapshot 2026-10-20.