ELEVENLABSSPEECH-TO-TEXT38,964 SAMPLES / 30 DAYSLAST RUN NOV 17, 2026, 03:00 UTC
Scribe v3 Live speech-to-text benchmarks
Scribe v3 Live, hosted by ElevenLabs (Eleven Labs, 11Labs), measures mean 98 ms time to final segment (10th of 30) and 2.0% word error rate (3rd of 31) among STT systems. Results cover the last 30 days. Last measured .
Coval has active speech-to-text measurements for Scribe v3 Live, created by ElevenLabs and served through ElevenLabs's own API.
- Time to Final Segment#10 / 30
- 98ms
- Word Error Rate#3 / 31
- 2%
- Time to First Token#9 / 27
- 1254ms
Overview
Scribe v3 Live is tested every day on fixed public audio — clean, accented, noisy, reverberant, far-field, clipped and phone-codec speech — for transcription accuracy and streaming latency.
Technical specifications
- Made by
- ElevenLabs
- Hosted by
- ElevenLabs
- Source
- Official API
- Licensing
- Proprietary
- Deployment
- Cloud
- Region
- US
- Features
- Keyterm biasing, Multilingual, VAD
How Scribe v3 Live ranks
Full STT dashboard- #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
Show all 30 modelsShow fewer
- #17Whisper Large v3Dedicated inference. Shared endpoints serve many customers on the same infrastructure, while dedicated endpoints run on hardware reserved for a single customer.125 ms
- #9Qwen3 ASR 1.7bDedicated inference. Shared endpoints serve many customers on the same infrastructure, while dedicated endpoints run on hardware reserved for a single customer.2.7%
Show all 31 modelsShow fewer
- #20Whisper Large v3Dedicated inference. Shared endpoints serve many customers on the same infrastructure, while dedicated endpoints run on hardware reserved for a single customer.4.1%
- #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
Show all 27 modelsShow fewer
| # | Model | Host | TTFS | WER | TTFT | Samples |
|---|---|---|---|---|---|---|
| 1 | Saidwire En 2026 10 30 | Saidwire | 29 ms | 768 ms | 9,441 | |
| 2 | Qwen3 ASR 1.7b | Baseten | 35 ms | 943 ms | 13,679 | |
| 3 | STT RT v5 | Soniox | 51 ms | 1529 ms | 27,358 | |
| 4 | Qwen3 ASR Fast | Nari Labs | 53 ms | 1757 ms | 27,361 | |
| 5 | STT Rt v6 | Soniox | 59 ms | 1893 ms | 12,188 | |
| 6 | STT 1 | Inworld AI | 70 ms | 994 ms | 27,333 | |
| 7 | Nova 3 | Deepgram | 95 ms | 1435 ms | 24,081 | |
| 8 | Nova 2 | Deepgram | 97 ms | 1432 ms | 24,046 | |
| 9 | Realtime Global | Reson8 | 98 ms | 1658 ms | 27,361 | |
| 10 | Scribe v3 Live | ElevenLabs | 98 ms | 1254 ms | 5,781 | |
| 11 | Universal 3.5 Pro | AssemblyAI | 104 ms | 1013 ms | 27,361 | |
| 12 | Universal 3.6 Pro | AssemblyAI | 110 ms | 1073 ms | 27,360 | |
| 13 | Flux Multilingual | Deepgram | 110 ms | 1173 ms | 24,079 | |
| 14 | Flux | Deepgram | 112 ms | 1095 ms | 24,085 | |
| 15 | Scribe | Zoom | 112 ms | — | 6,654 | |
| 16 | Scribe v2 Realtime | ElevenLabs | 116 ms | 2126 ms | 27,356 | |
| 17 | Whisper Large v3 | Baseten | 125 ms | 920 ms | 13,221 | |
| 18 | Ink 2 | Cartesia | 141 ms | 2030 ms | 18,412 | |
| 19 | Grok STT | SpaceXAI | 163 ms | — | 27,343 | |
| 20 | Pulse | Smallest | 169 ms | 1953 ms | 27,358 | |
| 21 | resonant-1 | Reson8 | 210 ms | — | 27,353 | |
| 22 | Linden 1 | Speechmatics | 229 ms | 1503 ms | 27,338 | |
| 23 | Scribe Pro v2.1 | Zoom | 241 ms | — | 6,823 | |
| 24 | Default | Gradium | 284 ms | 2074 ms | 27,354 | |
| 25 | Daytona STT | Guava | 348 ms | 5609 ms | 13,376 | |
| 26 | Gemini 3.5 Transcribe Live | Gemini | 355 ms | 1520 ms | 27,282 | |
| 27 | Solaria 1 | Gladia | 390 ms | 1534 ms | 27,318 | |
| 28 | GPT Realtime Whisper | OpenAI | 566 ms | 1813 ms | 26,878 | |
| 29 | Chirp 3 | 842 ms | 6035 ms | 27,311 | ||
| 30 | Chirp 2 | 909 ms | 6101 ms | 27,359 | ||
| — | Universal Streaming | AssemblyAI | — | 1511 ms | 27,330 |
Highest relative placement: 3rd of 31 on Word Error Rate.
Latency vs accuracy
Where the errors come from
- Scribe v3 Live2.0%
Averages and tail latency
Averages hide slow outliers — these are the distributions behind each figure.
- Time to Final Segmentp50 90 ms · Max 14694 ms
- Time to First Tokenp50 1285 ms · Max 4834 ms
- Scribe v3 Livep50 0.0% · Max 50.0%
| Metric | Average | p25 | p50 | p75 | p90 | p95 | Max | Samples |
|---|---|---|---|---|---|---|---|---|
| Time to Final Segment | 98 ms | 77 ms | 90 ms | 106 ms | 123 ms | 137 ms | 14694 ms | 5,781 |
| Word Error Rate | 2.0% | 0.0% | 0.0% | 3.1% | 8.3% | 11.8% | 50.0% | 5,781 |
| Time to First Token | 1254 ms | 968 ms | 1285 ms | 1582 ms | 1964 ms | 2269 ms | 4834 ms | 5,781 |
Last 30 days
Daily medians from the same measurement runs · gaps are days without qualifying runs.
Time to Final Segment by dataset
- WildASR phone codec94 ms
- WildASR accents95 ms
- PipeCat (production)95 ms
- WildASR noise gaps95 ms
- WildASR far-field99 ms
- WildASR clipping101 ms
- WildASR clean108 ms
| Dataset | TTFS | Samples |
|---|---|---|
| ProductionPipeCat | 95 ms | 3,049 |
| AccentsWildASR | 95 ms | 303 |
| CleanWildASR | 108 ms | 1,216 |
| ClippingWildASR | 101 ms | 304 |
| Far-fieldWildASR | 99 ms | 302 |
| Noise gapsWildASR | 95 ms | 304 |
| Phone codecWildASR | 94 ms | 303 |
Strongest condition: WildASR phone codec at 94 ms · weakest: WildASR clean at 108 ms.
How fast is Scribe v3 Live?
On ElevenLabs, Scribe v3 Live measures mean 98 ms time to final segment (10th of 30) and mean 1254 ms time to first token (9th of 27). Last measured 2026-11-17.
How accurate is Scribe v3 Live?
On ElevenLabs, Scribe v3 Live measures 2.0% word error rate (3rd of 31). Last measured 2026-11-17.
Who hosts Scribe v3 Live?
Scribe v3 Live is created by ElevenLabs and served by ElevenLabs. Coval measures each hosted endpoint separately.
Results are re-measured daily using fixed datasets and reported over a rolling 30-day window. 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.