Weyoto Voice is a modular Flask application that turns text into high-quality, realistic speech using OpenAI’s gpt-4o-mini-tts model. Users can customize tone using natural language instructions and generate downloadable MP3s in seconds — no microphone, no editing.

Built with a clean monolithic structure (no database), this is a perfect starter project or production-ready base for voice automation, storytelling tools, or creative audio projects.

• 🎤 AI-powered text-to-speech using OpenAI's streaming TTS API
• 🧠 Tone customization via natural instructions (e.g., “sound terrifying”)
• 🎧 Multiple voices: Coral, River, Shimmer, Onyx, etc.
• ⚡ Fast streaming audio generation
• 📅 Saves output as .mp3 files in static/audio/
• 🧱 Modular monolithic Flask structure
• 🔓 No database or auth required (MVP-friendly)
• 🧑‍💻 Simple HTML + JS frontend # Update

weyoto-voice/
├── app.py
├── config.py
├── extensions.py
├── modules/
│   └── tts/
│       ├── __init__.py
│       ├── logic.py          # TTS generation logic
│       ├── routes.py         # Flask routes for UI & API
│       └── templates/
│           └── index.html    # Basic frontend UI
├── app/
│   └── static/  
            └── audio/       # Saved MP3 files
├── requirements.txt
├── .gitignore
└── README.md

git clone https://github.com/your-username/weyoto-voice.git
cd weyoto-voice
python -m venv venv
source venv/bin/activate  # or venv\Scripts\activate on Windows
pip install -r requirements.txt

export OPENAI_API_KEY=your-openai-key  # macOS/Linux
# OR
$env:OPENAI_API_KEY="your-openai-key"  # Windows PowerShell

python app.py

Visit http://localhost:5000/tts in your browser.

POST /tts/generate-voice Content-Type: application/json

{
  "text": "The sailor was sailing at night and the waves were blowing hard. He decided to stop and not sail for the night",
  "voice": "coral",
  "instructions": "The voice should sound terrifying"
}

Response: downloadable MP3 file of the voiceover.

• Compatible with Render, Railway, or any VPS

• MP3 files are stored locally in static/audio/ — consider cleanup or time-based deletion

• Optional enhancements include:

  • Adding Tailwind for styling
  • Integrating with Wasabi/S3 for permanent storage
  • Adding auth or usage tracking later

• Auto-delete old audio files after 10 minutes
• Add reusable voice templates (presets)
• Add background music overlays
• Deploy at voice.weyoto.com
• Add Stripe/Paystack billing for pro usage
• Support streaming playback without saving

MIT License — free to use, extend, or commercialize.

Built with love by Deba, creator of Weyoto and builder of AI tools that actually work. For contributions or feedback, open an issue or pull request.

🧠 “Your words. A new voice.” — Weyoto Voice