Vals AI VoiceCodeBench: leaderboard
VoiceCodeBench (Vals AI) — speech-to-text on English workplace speech that carries exact structured values (URLs, email addresses, file paths, shell commands); Task Success Rate is the share of recordings where every target entity came back exactly right.
Metric: Task Success Rate (%). Source: www.vals.ai. Saturation forecast: Around January 2027. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Whisper Large-v3 | 54.33 | #1272 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=vals-ai-voicecodebench · How It Works · Data refreshed daily, snapshot 2026-10-21.