Documentation

Getting Started

Welcome to UVAI.io! This guide will help you integrate our video intelligence API into your application.

Quick Start

# Install the SDK
pip install uvai-python

# Or use npm
npm install @uvai/sdk

Authentication

All API requests require an API key. Include it in the header:

curl -X GET https://uvai.io/api/v1/health \
  -H "X-API-Key: your_api_key_here"

API Endpoints

GET /api/v1/health

Check API health and connection status.

POST /api/v1/transcript-action

Process a video and extract transcript with events.

POST /api/v1/generate

Transform video into deployable infrastructure (Revenue Pipeline).

POST /api/video-to-software/by-category

Auto-discover videos by category and process them.

Transcription API

Extract word-perfect transcripts from any video with speaker identification.

Request

POST /api/v1/transcript-action

{
  "video_url": "https://youtube.com/watch?v=example",
  "language": "en",
  "extract_events": true,
  "dispatch_agents": false
}

Response

{
  "transcript": "Full transcript text...",
  "segments": [
    {
      "start": 0.0,
      "end": 5.2,
      "text": "Welcome to the video",
      "speaker": "Speaker 1"
    }
  ],
  "events": [...],
  "metadata": {
    "duration": 1234,
    "language": "en",
    "word_count": 5678
  }
}

Event Extraction

Automatically identify action items, key moments, and structured events.

{
  "events": [
    {
      "type": "action_item",
      "timestamp": 125.4,
      "content": "Set up the development environment",
      "confidence": 0.95
    },
    {
      "type": "key_moment",
      "timestamp": 340.0,
      "content": "Important announcement about feature release",
      "confidence": 0.88
    }
  ]
}

Agent Execution

Dispatch AI agents to take real actions based on video content.

Supported Agents

Code Generator: Transform tutorials into working code
Ticket Creator: Create issues from action items
Document Generator: Create documentation from explanations
Workflow Trigger: Execute custom workflows via webhooks