OmniACBench - Script Fidelity WER: leaderboard
Metric: Word error rate as printed (human reference 0.05; the same models reading the script alone score 0.09-0.15) of the generated speech, transcribed by Whisper-large-v3, against the script, on OmniACBench's 3,559 synthetic instances (spoken instruction, text script and image; the model must read the script aloud with the delivery the context implies); lower is better. Source: arxiv.org. Saturation forecast: Around April 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3 Omni 30B A3B Instruct | 2.14 | #362 |
| 2 | Qwen2.5-Omni-7B | 4.15 | #601 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=omniacbench-script-fidelity-wer · How It Works · Data refreshed daily, snapshot 2026-10-11.