GoogleElevenLabsAudio matchupSpeech to text
Gemini 3.5 Transcribevs.Scribe v2
Which model is better?
Scribe v2 does better in independent tests, while Gemini 3.5 Transcribe is more popular.
Pick Gemini 3.5 Transcribe for
Pick Scribe v2 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”.
Ocular AI Converse-STT Version 2
Scribe v2 slightly aheadScribe v2 is clearly ahead in 0 of 2 subtests and slightly in 2, Gemini 3.5 Transcribe in none.
Ocular AI Converse-STT Version 2 pooled character error rate (CER)
lower is better, as of October 22, 2026
Ocular Converse-STT Version 2 pooled WER
lower is better, as of October 17, 2026
Source · latest October 22, 2026
Appen MedTerm-90
Scribe v2 aheadAppen MedTerm-90 overall clinical entity recall
95 % ranges 79.5 % to 81.8 % and 82.6 % to 84.1 %, 30,017 tasks, as of October 16, 2026
Source · latest October 16, 2026
Artificial Analysis
Scribe v2 slightly aheadArtificial Analysis AA-WER (non-streaming)
lower is better, as of October 15, 2026
Source · latest October 15, 2026
All data side by side
Prices are verbatim from the provider, because audio models bill per minute, character or token.
| Gemini 3.5 Transcribe | Scribe v2 | |
|---|---|---|
| Task | Speech to text | Speech to text |
| Developer | ElevenLabs | |
| Status | Active | Active |
| Weights | proprietary | proprietary |
| Runs locally | no | no |
| Price | $2 per 1M audio tokens ($0.003 per audio minute) | n/a |
| Popularity (out of 100) | 52 | 39 |