Detect Deepfake

note

Detect API is currently in beta. Please contact us if you would like to be added to the beta program.

Submit an audio file for deepfake detection.

HTTP Request

POST https://app.resemble.ai/api/v2/detect

Request

Parameter	Type	Description
url	Required	The URL of the audio file to be analyzed for deepfake detection. The audio file should be in WAV or MP3 format. Provide a valid HTTPS URL pointing to the audio file.
callback_url	Optional	A URL to sent the results to when analysis is completed.

Deprecated Parameters

Parameter	Type	Description
frame_length	Optional	The length of the frames in seconds used for analysis. The default value is 2 seconds. You can adjust this value based on the characteristics of your audio data and the desired analysis granularity. The minimum value is 1 and the maximum value is 4.
sensitivity	Optional	Tune the aggressiveness towards fake or real detection. Increasing this value towards 1 will more aggressively label clips as fake and may result in more false positives. (maximum: 1)

Callback URL

If you've provided a callback_url when submitted an audio, you will receive a POST request to the callback_url with the following payload after it has been analyzed:

{
  "success": true,
  "item": {
    "uuid": <string>,
    "metrics": {
      "label": "fake",
      "score": ["0.9"]
    }
  }
}

HTTP Response

{
    "success": true,
    "item": {
        "uuid": <string>,
        "metrics": null
    }
}