Text-to-Video Generator

Generate 10-minute YouTube videos with voice narration from text input, now with AI image generation and modification capabilities.

Features

• OpenAI TTS - Premium human-like voice narration (with free Windows SAPI fallback)
• NEW: Direct YouTube Upload - Automatically publish videos to your YouTube channel
• AI Image Generation & Modification - Generate and modify images using AI prompts
• Automatic text segmentation for optimal video length
• Visual text overlay on video frames
• MP4 output optimized for YouTube
• Multiple AI image providers (Pollinations, Hugging Face)
• Smart prompt enhancement for better image quality
• OAuth 2.0 integration with YouTube API
• Customizable video metadata (title, description, tags, privacy)

Installation

npm install

Usage

Web Interface (Recommended)

npm run web

Then open http://localhost:8081 in your browser

Command Line

npm start

Requirements

• Node.js 16+
• FFmpeg installed on system
• OpenAI API key (optional, for premium TTS - see OPENAI-SETUP.md)
• Google Cloud Console account (for YouTube uploads)
• YouTube channel (for uploads)

Configuration

OpenAI TTS Setup (Recommended)

1. Follow the detailed setup guide in OPENAI-SETUP.md
2. Set your OPENAI_API_KEY environment variable
3. Enjoy premium human-like voices (onyx, alloy, echo, fable, nova, shimmer)
4. Falls back to free Windows SAPI if not configured

YouTube Upload Setup

1. Follow the detailed setup guide in YOUTUBE-SETUP.md
2. Create youtube-config.json with your Google OAuth credentials
3. Authorize the application through the web interface
4. Enable "Upload to YouTube" option when generating videos

Quick Setup:

# Copy example config
cp youtube-config.example.json youtube-config.json

# Edit with your Google Cloud credentials
# Then authorize through web interface

Output

Videos are saved to the output/ directory as MP4 files ready for YouTube upload.