A modern web interface for Gujarati Text-to-Speech synthesis using the F5-TTS model.

• Text-to-Speech Conversion: Convert Gujarati text to natural-sounding speech
• Advanced Generation Parameters: Customize speech generation with adjustable parameters
• Dark/Light Theme: Toggle between dark and light mode for comfortable viewing
• Audio History: View, play, and download previously generated audio
• Responsive Design: Works on desktop and mobile devices
• User Authentication: Basic login system to protect the application

• Backend: Flask (Python)
• Frontend: HTML, CSS, JavaScript
• TTS Engine: F5-TTS model via Gradio API
• Audio Processing: Generated WAV files with customizable settings

• Python 3.8+
• Gradio API running locally on port 7860
• The F5-TTS model files for Gujarati

1. Clone the repository:

git clone https://github.com/Ahir7/f5-tts-web-app.git
cd f5-tts-web-app

2. Create a virtual environment and activate it:

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

3. Install dependencies:

pip install -r requirements.txt

4. Configure the model paths in app.py: Update the LANGUAGES dictionary with your local model paths.

1. Start the Gradio API for the TTS model on port 7860.

2. Run the Flask application:

python app.py

3. Access the web interface at http://localhost:5000

1. Login: Use the default credentials (admin/admin123) or update them in the app
2. Enter Text: Type or paste Gujarati text in the text area
3. Adjust Parameters (optional):
   • NFE Step: Adjust the number of function evaluations (default: 32)
   • Speed: Control speech speed (0.5-2.0)
   • Random Seed: Set for consistent output (or -1 for random)
   • Remove Silence: Toggle to trim silence
   • Use EMA: Toggle Exponential Moving Average usage
4. Generate: Click "Generate Speech" to process the text
5. Listen & Download: Play the audio in the browser or download it
6. View History: Access previously generated audio files

• app.py: Main Flask application
• templates/: HTML templates
  • index.html: Main TTS interface
  • login.html: Login page
  • history.html: Audio history page
• static/: Static assets
  • css/style.css: Application styling
  • js/script.js: Client-side functionality
  • audio/: Generated audio files

• Adding Languages: Add new language configurations to the LANGUAGES dictionary in app.py
• Styling: Modify the static/css/style.css file to change the appearance
• Users: Update the users dictionary in app.py to manage authentication

MIT License

• F5-TTS Model Team for the text-to-speech technology
• Contributors and maintainers of the original F5-TTS project

• Database integration for user management
• Multiple language support
• Batch processing capabilities
• API key authentication
• Customizable voice profiles