Full leaderboard and methodology for Zoom Scribe streaming STT benchmark
ArtificialAnlys · x · 2026-11-20
Artificial Analysis published full results and methodology for the Zoom Scribe V2.1 Pro Live benchmark. The AA-WER Streaming index uses 8 hours of real-world audio from AA-AgentTalk (50%), VoxPopuli (25%) and Earnings22 (25%), measuring streaming word error rate, latency to final transcript after speech end, and price per 1,000 minutes across 54 models.
Related event: Zoom Scribe Tops Streaming ASR Benchmark at 2.4% WER, but Lags in Latency(4 posts)→
More from Models
- OpenAI: keep conversation prefix unchanged when using GPT-6.2 in Responses API — OpenAIDevs · 2026-11-20
- Engineer's first take on Hy4-Flash: clean code out of the box, hours of heavy use cost $0.10 — crusaderky · 2026-11-20
- Zoom Scribe pricing: $0.40/hour, between Microsoft and ElevenLabs — ArtificialAnlys · 2026-11-20
- Zoom Scribe V2.1 Pro Live tops streaming STT accuracy at 2.4% WER — ArtificialAnlys · 2026-11-20
- Gemini API adds tool search preview: defer tool loading so agents skip sending full tool libraries — rseroter · 2026-11-20
- scaling01 revises Google verdict: now just 1-2 months behind frontier, not 3-4 — scaling01 · 2026-11-20