Provider pricing
Speechmatics AI API pricing
Speechmatics' Ursa speech recognition engine delivers state-of-the-art accuracy across 50+ languages and accents, including hard-to-handle dialects that trip up Whisper. Real-time streaming, speaker diarization, and on-prem deployment make it a go-to for broadcast, contact centers, and regulated transcription workflows.
Founded2006
HeadquartersCambridge, UK
Tracked models2
Best for
Broadcasters, contact centers, and enterprises that need top-tier transcription accuracy across global accents.
Strengths
- Industry-leading accuracy on accented English and code-switched audio
- 50+ languages with real-time and batch streaming endpoints
- Speaker diarization, custom vocabulary, and entity formatting
- On-prem and self-hosted deployments for media and regulated sectors
Tracked Speechmatics models
| Model | Input $/M | Output $/M | Context | Tier | Best for |
|---|---|---|---|---|---|
| Speechmatics Standard | $0.300 | $0 | 0 | Budget | STT: $0.30 per audio hour — 50+ languages with diarization |
| Speechmatics Enhanced | $1.04 | $0 | 0 | Flagship | Best-in-class STT: $1.04 per audio hour, on-prem available |