Developer Docs

Ship voice security
in minutes.

One API gives you access to self-trained SOTA deep learning models. deepfake detection, intent analysis, speaker verification, and audio intelligence across 50+ languages. >99% accuracy on the "In-the-Wild" dataset. SDKs for Python and Node.js.

REST APIWebSocket StreamingPython SDKNode.js SDK50+ Languages>99% AccuracyVoicemail & IVR99.9% Uptime SLA

Quick Start

Install an SDK and make your first detection call in under 5 minutes.

1

Install the SDK

pip install vocos# or: npm install @vocos/node
2

Get your API key

Sign up at dashboard.vocos.ai and generate a live API key. Keys start with vcs_live_.

3

Make your first call

from vocos import Vocos
result = Vocos(api_key="vcs_live_...").detect("audio.wav")

API Reference

Three products, one API key. Select a product to see code examples.

1from vocos import Vocos
2
3client = Vocos(api_key="vcs_live_...")
4
5# Detect deepfakes in a single audio file
6result = client.detect("call_recording.wav")
7
8print(result.verdict)      # "synthetic" | "authentic"
9print(result.confidence)   # 0.994
10print(result.latency_ms)   # 23
11print(result.artifacts)    # ["formant_smearing", ...]
12print(result.language)     # "en-US"

Endpoints

Base URL: https://api.vocos.ai

POST/v1/detect
POST/v1/detect/intent
POST/v1/detect/intelligence
POST/v1/shield/stream
POST/v1/shield/configure
POST/v1/verify/enroll
POST/v1/verify/authenticate
GET/v1/languages
Authentication

Bearer Token

All requests use Bearer authentication. Include your API key in the Authorization header.

Authorization: Bearer vcs_live_...
Rate Limits

Generous defaults

  • Detect1,000 req/min
  • Shield100 concurrent streams
  • Verify500 req/min

Enterprise plans include custom rate limits.

Ready to integrate?

Get an API key and start detecting synthetic speech in under 5 minutes.

Get API KeyTalk to Sales