Transcribe audio
Speechmatics
Speech-to-text API that transcribes recorded files and live audio streams into text with timings and speaker labels.
At a glance
- Provider
- Speechmatics
- Category
- Transcribe audio
- Type
- Hosted API
- Pricing model
- Usage-based credits, billed per second at an hourly rate2
- Free tier
- $100 credit grant, no card3
- Authentication
- API key as bearer token; temporary JWTs for client-side use5
- Connect via
- REST API (batch), WebSocket (realtime), Python, JavaScript, .NET and Rust SDKs13
- Read from the provider's pages
- 2026-11-04
Pricing
- Pricing model: Pay as you go, credit packs or subscription; credits at $1 each, billed per second
- Free tier: $100 credit grant for new sign-ups
- Card required for free tier: No
- Official MCP server: No
- Address or package: Not published
- Who creates the account: A person signs up in the portal; agent self-sign-up is not stated
- Where to get a key: Portal, Settings > API Keys; Management API with a management token
- Other requirements: Enterprise customers may need to contact support for keys; no card for the free plan
- REST API:
https://eu1.asr.api.speechmatics.com/v2(alsous1andau1) - MCP server: Not published
- OpenAPI spec: Not published
- Batch job creation: 10 new jobs per second recommended; HTTP 429 beyond that
- Batch status requests: 50 per second recommended
- Concurrent realtime sessions: Free 2, Pro 50, Enterprise custom
- Data location: EU1, US1 or AU1, chosen by the endpoint called
- Retention: Batch: stored 7 days then deleted; realtime: not stored
- Used for training: Only if you opt in to the model-training discount; Enterprise says no logging by default
Read from Speechmatics’s own pages on 2026-11-04. Every number links to the page a value came from. For agents, the same facts as markdown: /tools/transcribe-audio/speechmatics.md