NeuroVox - Advanced Text-to-Speech Studio

NeuroVox is a high-performance, self-hosted text-to-speech (TTS) application powered by neural networks. It provides a modern web interface to generate natural-sounding speech from text using the Piper TTS engine.

🚀 Features

• High-Quality Neural Voices: Utilizes advanced ONNX-based models for human-like speech synthesis.
• Multi-Language Support: Ready for English (US) and Hindi, with support for adding many more.
• History & Management: Automatically saves generation history, allowing you to replay and download past audio.
• Modern UI: A responsive, glass-morphism inspired interface with dark mode elements.
• Dockerized: Easy deployment with Docker and Docker Compose.
• Offline Capable: Runs entirely locally; no internet connection required for generation after initial setup.

💡 Use Cases

NeuroVox is versatile and can be used for various applications:

1. Content Creation: Generate voiceovers for YouTube videos, podcasts, and social media content without expensive voice actors.
2. Accessibility: Create audio versions of articles, blog posts, and documents for visually impaired users.
3. E-Learning: Produce consistent and clear narration for educational courses and training materials.
4. Prototyping: Quickly generate placeholder audio for game development, animations, or app interfaces.
5. Personal Assistant: Integrate into home automation systems for custom voice notifications.

🛠️ Deployment Steps

Prerequisites

• Docker and Docker Compose installed on your system.

Quick Start (Recommended)

We provide a convenience script to handle the build and startup process automatically.

1. Clone the repository:

   git clone <repository-url>
   cd Text2Speech

2. Run the start script:

   ./start.sh

   This will build the necessary images and start the application.

3. Access the application: Open your browser and navigate to http://localhost:5000.

Advanced Options

• Run in background:

  ./start.sh -d

• Force rebuild base image:

  ./start.sh --build-base

Manual Docker Deployment

If you prefer to run Docker commands manually:

1. Build the base image:

   docker build -f Dockerfile.base -t text2speech-base .

2. Start with Docker Compose:

   docker-compose up --build

Local Python Setup (No Docker)

If you wish to run it directly on your machine:

1. Install Dependencies:

   sudo apt-get install espeak-ng
   pip install -r requirements.txt

2. Run the App:

   python app.py

🗣️ Adding Voice Models

NeuroVox uses Piper TTS models (.onnx and .json files).

1. Download Models: Visit the Piper Voices Hugging Face repository to find models.

   Example commands to download voices:

   # Install piper-tts to use the download utility
   pip install piper-tts
   
   # Download specific voices
   python3 -m piper.download_voices en_US-amy-medium
   python3 -m piper.download_voices en_US-libritts_r-medium

2. Install: Place the downloaded .onnx and .onnx.json files into the voice_models/ directory.

3. Restart: Restart the application (or Docker container) to detect the new voices.

📄 License

MIT License