SpeechParaling-Bench (English) - Situational Adaptation: leaderboard
Metric: Pairwise score (%) on the situational adaptation task (answer a spoken query with fitting content and speaking style), English subset: a Gemini 3 Pro judge compares each spoken response with the fixed baseline's (Gemini 2.5 Flash Audio; ties count half), and the baseline's own row is its strength-weighted average against the candidates (the paper's Eq. 3); higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT Audio (2025-08-28) | 57.68 |
| 2 | Gemini 2.5 Flash Audio | 52.37 |
| 3 | Qwen3-Omni-Realtime | 48.57 |
| 4 | Doubao Realtime Voice Model | 46.07 |
| 5 | Qwen3-Omni-Flash-2025-12-01 | 20.18 |
Interactive version: theaggregate.ai/benchmark?slug=speechparaling-bench-english-situational-adaptation · How It Works · Data refreshed daily, snapshot 2026-10-07.