OllamaVoice

OllamaVoice

An AI-powered research and video generation tool that combines Ollama's language capabilities with OuteTTS for high-quality text-to-speech generation.

Features

• Text-to-Speech generation using OuteTTS
• Research capabilities powered by Ollama
• Video generation with AI narration
• Support for multiple languages (English, Japanese, Korean, Chinese)
• Real-time audio generation
• Beautiful web interface

Requirements

• Python 3.12+
• Ollama running locally
• FFmpeg for video processing
• PyTorch (with CUDA support recommended)

Installation

1. Clone the repository:

git clone https://github.com/yourusername/OllamaVoice.git
cd OllamaVoice

1. Create and activate virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

1. Install dependencies:

pip install -r requirements.txt

1. Create necessary directories:

mkdir -p static/temp static/videos

1. Create .env file with your configuration:

OLLAMA_BASE_URL=http://localhost:11434
OLLAMA_MODEL=mistral

Usage

1. Start the server:

uvicorn main:app --reload

1. Open your browser and navigate to `http://localhost:8000`

2. Use the interface to:

   • Generate TTS audio
   • Research topics
   • Create AI-narrated videos

Architecture

• FastAPI backend
• OuteTTS for text-to-speech
• Ollama for AI research
• FFmpeg for video processing
• Modern web interface

Documentation

See the `docs` directory for detailed documentation:

• Token Handling

License

MIT License