Speech Revolutions vs ElevenLabs
ElevenLabs is a voice-generation company whose Scribe models add speech-to-text to its text-to-speech and voice-agent products. We build transcription only, and the difference that shows up in measurement is speaker diarization.
Where we differ measurably is speaker diarization — telling apart who said what — and that gap is large enough to decide the question for meetings, interviews and contact-centre audio.
Diarization error rate
Lower is better. Overlap-aware at a 0.25 s collar, on public datasets, measured 2026-07-12 against ElevenLabs (Scribe V2).
| Dataset | Speech Revolutions | ElevenLabs |
|---|---|---|
| AMI-SDM | 9.7% | 33.7% |
| NotSoFar | 10.9% | 24.4% |
| DiPCo | 26.7% | 31.6% |
The harness that produced these is public, and so are the datasets — so this is a number you can reproduce or contest rather than take on trust. Full benchmarks and methodology.
Differences that matter in practice
| Speech Revolutions | ElevenLabs | |
|---|---|---|
| Product scope | Transcription, diarization and timestamps. | Speech-to-text alongside text-to-speech, voice cloning and agents. |
| Real-time | Not offered. Batch only. | A separate Scribe Realtime model. |
| Diarization | On by default, with speaker turns already grouped. | diarize=true adds a speaker_id to each word. |
| Response words | Words only. | Words interleaved with spacing tokens to filter out. |
Pricing
We charge $0.003 per audio minute — $0.18 an hour — with speaker diarization, word timestamps, punctuation and keyword biasing included rather than priced separately.
We deliberately do not quote ElevenLabs's prices here. They are tiered, they change without notice, and a stale number on a page like this is worse than no number at all. The comparison worth making is the effective cost of one audio hour with the features you actually need switched on — check theirs against ours directly.
What switching involves
Auth moves from xi-api-key to X-API-Key, and the single multipart POST becomes one transcribe() call in the SDK. Your diarize flag works unchanged, result.text maps directly, and the code that filtered out spacing tokens can go.
The migration guide maps every field, side by side, in four languages.
More: diarization measured across five APIs · ElevenLabs alternatives.