API Reference

Base URL, authentication, request conventions, and the OpenAPI description of every endpoint.

The DetectifAI REST API checks recordings for AI-generated speech, and enrolls and verifies speakers by voice. Every endpoint has its own page, generated from the OpenAPI description, with examples in cURL, Python and JavaScript.

Base URL

Example
https://platform.detectif.ai/v1

Authentication

Send your API key as a bearer token on every request:

Shell
curl https://platform.detectif.ai/v1/speakers \
  -H "Authorization: Bearer $DETECTIF_API_KEY"

See Authentication.

Requests and responses

  • Recordings are sent as multipart/form-data, with the audio in the file field (up to 100 MB). See Audio and limits.
  • Every response is JSON and includes a request_id, which is also sent in the X-Request-Id header.
  • Send an Idempotency-Key header with each upload and reuse it when you retry. A retried request returns the same result instead of running again.
  • Errors share one shape with a stable error.code. See Errors.

OpenAPI and Postman

The full description of the API, for code generators and API tools:

Endpoints

Deepfake detection

Method Path Page
POST /v1/deepfake/detect Detect AI-generated speech

Speakers

Method Path Page
POST /v1/speakers/enroll Enroll a voice sample
POST /v1/speakers/verify Verify a speaker
GET /v1/speakers List speakers
GET /v1/speakers/{speaker_id} Get a speaker
DELETE /v1/speakers/{speaker_id} Delete a speaker

Results

Method Path Page
GET /v1/operations/{operation_id} Get a result
POST /v1/operations/{operation_id}/feedback Say whether a result was right

Account

Method Path Page
GET /v1/credits Get your balance
GET /v1/usage List your usage

System

Method Path Page
GET /v1/health Check service health