AssemblyAI

API

Speech-to-text API for recorded files and live audio streams

Price: Free credits, then paid
Access: API key

About

AssemblyAI provides speech-to-text models for recorded files and live streams, with add-ons such as speaker labels, PII redaction and translation. Universal-3.5 Pro costs $0.21 per hour, and new accounts get $50 in credit. An EU endpoint keeps data in the EU.

What you can do with it

Transcribe recorded audio or video files with speaker labels
Stream live audio over WebSocket for real-time captions and voice apps
Run a voice agent over one WebSocket with the Voice Agent API

Get started

Sign up and copy your API key from the dashboard ($50 free credit)
Export it as ASSEMBLYAI_API_KEY

Example

curl -X POST https://api.assemblyai.com/v2/transcript \
  -H "authorization: $ASSEMBLYAI_API_KEY" \
  -H "content-type: application/json" \
  -d '{"audio_url": "https://assembly.ai/wildfires.mp3", "speaker_labels": true}'
# Poll GET https://api.assemblyai.com/v2/transcript/<id> until status is "completed"

Details

Hosting: Hosted service, Self-hosted
Available in: Worldwide
Official SDKs: Python, JavaScript/TypeScript
MCP server: None