AuthGuard API

The AuthGuard REST API allows you to programmatically access our enterprise-grade AI detection models. Integrate deepfake detection, voice cloning analysis, and AI-text classification directly into your own applications.

Your API Key

Use this secret key to authenticate your requests. Do not share it.

Authentication

Authenticate your API requests by including your secret API key in the Authorization HTTP header. All API requests must be made over HTTPS.

cURL

curl -H "Authorization: Bearer sk_test_ag_XXXXXX" \
     https://api.authguard.ai/v1/models

Detect Voice (Audio)

Analyze an audio file to determine if it was synthesized using AI voice cloning technology (e.g., ElevenLabs, Bark).

POST /v1/detect/voice

Parameter	Type	Description
file	file	The audio file to analyze. Supported formats: `.wav`, `.mp3`, `.ogg`. Max size: 25MB.
mode	string	Optional. Set to `fast` for speed or `deep` for highest accuracy.

Python

Node.js

import requests

url = "https://api.authguard.ai/v1/detect/voice"
headers = { "Authorization": "Bearer YOUR_API_KEY" }
files = { "file": open("suspect_audio.wav", "rb") }

response = requests.post(url, headers=headers, files=files)
print(response.json())

Response Example

{
  "id": "req_8x92nd10a",
  "status": "success",
  "analysis": {
    "is_ai_generated": true,
    "confidence_score": 99.4,
    "detected_model": "ElevenLabs",
    "artifacts": ["high frequency cutoff", "unnatural breath cadence"]
  }
}

Detect Image (Deepfake)

Analyze an image to detect manipulation, stable diffusion generation, or GAN synthesis.

POST /v1/detect/image

Works identically to the voice endpoint, but accepts .jpg, .png, and .webp formats.