Speechocean762 (Realtime Multi-Granularity Repeat) - Utterance Total: leaderboard
Metric: Pearson correlation (-1 to 1) with expert utterance total ratings on valid outputs from the 2,500-utterance Speechocean762 test split; GPT-4o Realtime zero-shot, temperature 0.6, multi-granularity repeat prompt/run; 47.76% of the corresponding utterance/word/phoneme items unscored and excluded from correlation; higher is better. Source: arxiv.org. 1 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o Realtime (Speechocean762 snapshot unspecified) | 0.46 |
Interactive version: theaggregate.ai/benchmark?slug=speechocean762-realtime-multi-granularity-repeat-utterance-total · How It Works · Data refreshed daily, snapshot 2026-10-12.