ElevenLabsAssemblyAIAudio matchupSpeech to text
Scribe v2 Realtimevs.Universal-3.6 Pro Realtime
Which model is better?
Scribe v2 Realtime and Universal-3.6 Pro Realtime each lead in different tests.
Pick Scribe v2 Realtime for
Pick Universal-3.6 Pro Realtime for
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
Coval
depends on the subtestUniversal-3.6 Pro Realtime is clearly ahead in 0 of 15 subtests and slightly in 5, Scribe v2 Realtime in 0 clearly and slightly in 4. 6 are level.
Coval Time to First Token (TTFT) maximum
lower is better, as of November 5, 2026
Coval Time to Final Segment (TTFS) maximum
lower is better, as of November 5, 2026
Coval AudioToFinal maximum
lower is better, as of November 5, 2026
Show all 15 subtests
| Subtest | Scribe v2 Realtime | Universal-3.6 Pro Realtime | Verdict |
|---|---|---|---|
| Coval Time to First Token (TTFT) maximum (lower is better) | 4,960 ms | 36,915 ms | Scribe v2 Realtime slightly ahead |
| Coval Time to Final Segment (TTFS) maximum (lower is better) | 5,714 ms | 27,919 ms | Scribe v2 Realtime slightly ahead |
| Coval AudioToFinal maximum (lower is better) | 16.82 s | 36.92 s | Scribe v2 Realtime slightly ahead |
| Coval RTF maximum (lower is better) | 1.96 | 4.10 | Scribe v2 Realtime slightly ahead |
| Coval Time to First Token (TTFT) mean (lower is better) | 2,150 ms | 1,077 ms | Universal-3.6 Pro Realtime slightly ahead |
| Coval pooled corpus Word Error Rate (WER) (lower is better) | 4.2 % | 2.2 % | Universal-3.6 Pro Realtime slightly ahead |
| Coval Time to First Token (TTFT) p50 (lower is better) | 2,111 ms | 1,118 ms | Universal-3.6 Pro Realtime slightly ahead |
| Coval Time to Final Segment (TTFS) p50 (lower is better) | 112 ms | 99 ms | Universal-3.6 Pro Realtime slightly ahead |
| Coval Time to Final Segment (TTFS) mean (lower is better) | 119 ms | 110 ms | Universal-3.6 Pro Realtime slightly ahead |
| Coval RTF mean (lower is better) | 1.02 | 1.02 | level |
| Coval AudioToFinal p50 (lower is better) | 8.45 s | 8.43 s | level |
| Coval AudioToFinal mean (lower is better) | 8.07 s | 8.06 s | level |
| Coval RTF p50 (lower is better) | 1.01 | 1.01 | level |
| Coval Word Error Rate (WER) per-sample maximum (lower is better) | 100.0 % | 100.0 % | level |
| Coval Word Error Rate (WER) per-sample p50 (lower is better) | 0.0 % | 0.0 % | level |
Source · latest November 5, 2026
All data side by side
Prices are verbatim from the provider, because audio models bill per minute, character or token.
| Scribe v2 Realtime | Universal-3.6 Pro Realtime | |
|---|---|---|
| Task | Speech to text | Speech to text |
| Developer | ElevenLabs | AssemblyAI |
| Status | Active | Active |
| Released | n/a | September 29, 2026 |
| Weights | proprietary | proprietary |
| Runs locally | no | no |
| Price | n/a | $0.45 per hour of audio; volume discounts available |
| Popularity (out of 100) | 29 | 0 |