WavBench - Explicit Acoustic Generation: leaderboard
Metric: Accuracy (%) of generated speech against the requested paralinguistic attribute, annotated by Gemini 3 Pro Preview, unweighted mean over ten attributes (accent, age, emotion, gender, language, pitch, speed, volume, audio, music); higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3 Omni 30B A3B Instruct | 62.03 | #362 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=wavbench-explicit-acoustic-generation · How It Works · Data refreshed daily, snapshot 2026-10-11.