Complete media analysis
Transcript, diarization, word and sentence timestamps, language, SRT, VTT, and structured JSON from one asynchronous job.
THE AUDIO INTELLIGENCE API
Media in.
Intelligence out.
One secure endpoint turns audio or video into transcripts, speaker labels, timestamps, subtitles, and structured JSON—processed on dedicated Apple silicon.
01 POST media→02 Job ID→03 Poll or webhook→04 JSON result
Queue-first reliability
Receive a job ID immediately, monitor state through the API, or accept a signed completion webhook in your application.
Short-lived by default
Choose immediate source deletion or encrypted retention from one hour to seven days. Results remain customer-scoped.
DEVELOPER WORKSPACE
Build and monitor
Your key stays in this browser tab.
New processing job
OR INGEST A PUBLIC HTTPS URL
Recent jobs
Connect an API key to view jobs.
CLEAR, USAGE-BASED PRICING
Pay for processed minutes.
Nothing else.
Start with a Square checkout pilot. Move to consolidated Square usage invoices when your integration is ready.
$24
one-time · 1,200 minutes
- Complete intelligence output
- API keys and dashboard
- Signed webhooks
- 7-day maximum retention
2¢
per processed minute
- Consolidated Square invoices
- Higher limits and priority queue
- Volume pricing available
- Integration support
PRIVACY & SECURITY
Your customers’ media is not our training data.
TLS protects every upload, download, webhook, and API request. Stored objects use platform encryption at rest.
API keys are stored as one-way hashes. Jobs and results are scoped to the authenticated customer.
Source files can be deleted immediately after processing or retained for a customer-selected window up to seven days.
Every completion event is signed with HMAC-SHA256 so receiving systems can verify it before acting.
Do not submit regulated or highly sensitive data until a written data-processing agreement and required compliance controls are in place. The service is an early commercial pilot; generated results should be reviewed.