PROVIDERLAST 30 DAYS
Zoom voice AI models and benchmarks
Zoom lists 2 STT models in Coval. Its fastest dated STT result is Scribe at mean 112 ms time to final segment (15th of 30) among STT systems, with 2.6% WER. Results cover the last 30 days. Last measured .
Coval measures voice models created or hosted by Zoom.
- Measured models
- 2
- STT
Overview
Every model below is measured daily on the same fixed inputs and ranked against the full field, never blended into a company score.
Model lineup
- Speech-to-Text
- Scribe, Scribe Pro v2.1
Speech-to-Text
Full STT dashboardRanked on Time to Final Segment against 30 measured models.
- #2Qwen3 ASR 1.7bDedicated inference. Shared endpoints serve many customers on the same infrastructure, while dedicated endpoints run on hardware reserved for a single customer.35 ms
- #2Whisper Large v3Dedicated inference. Shared endpoints serve many customers on the same infrastructure, while dedicated endpoints run on hardware reserved for a single customer.920 ms
- #3Qwen3 ASR 1.7bDedicated inference. Shared endpoints serve many customers on the same infrastructure, while dedicated endpoints run on hardware reserved for a single customer.943 ms
| Model | Host | TTFS | Rank |
|---|---|---|---|
| Scribe | Zoom | 112 ms | 15th of 30 |
| Scribe Pro v2.1 | Zoom | 241 ms | 23rd of 30 |
How fast are Zoom's STT models?
Scribe measures mean 112 ms time to final segment (15th of 30) among STT systems. Last measured 2026-11-14. Scribe Pro v2.1 measures mean 241 ms time to final segment (23rd of 30) among STT systems. Last measured 2026-11-14.
How accurate are Zoom's STT models?
Scribe Pro v2.1 measures 1.7% word error rate (1st of 31) among STT systems. Last measured 2026-11-14. Scribe measures 2.6% word error rate (8th of 31) among STT systems. Last measured 2026-11-14.
Which Zoom model is fastest?
Its fastest dated STT result is Scribe at mean 112 ms time to final segment (15th of 30) among STT systems, with 2.6% WER. Last measured 2026-11-14.
Same datasets, prompts and metric definitions for every model, measured by Coval’s open-source runner. Full methodology on the overview.
Evaluate your own voice agent
Use Coval to test your production configuration, prompts and calls—not only the public benchmark endpoints.