OpenAI & Azure TTS Demo in GitHub Codespaces

This project demonstrates how to convert text to speech using the TTS APIs from OpenAI and Azure. It has been reworked to provide a streamlined development experience in GitHub Codespaces with a modern file structure and improved diagnostics.

Project Structure

• tts_demo.py: Main Python script that reads text from text_to_convert.txt and generates speech audio.
  • For OpenAI TTS, the script outputs output.mp3.
  • For Azure TTS, the script outputs output_azure.mp3.
• text_to_convert.txt: Contains the text to convert into speech.
• .env: Environment file (should contain your API keys and region) that is loaded via Codespaces secrets.
• requirements.txt: Lists the Python dependencies.
• .devcontainer/: Contains the devcontainer.json for setting up your Codespace.
• diagnostics/: Contains scripts for verifying environment variable injection and Azure configuration (validate_env.py and validate_azure.py).
• .gitignore: Ensures that generated audio files, environment files, and cache directories are not committed.

Features

• Dual TTS Endpoints: Easily switch between OpenAI and Azure TTS using command-line options.
• Customizable Voices: Select voices using the command-line arguments (--voice for Azure, --voice-oai for OpenAI).
• Diagnostics Tools: Validate your environment setup with the scripts in the diagnostics folder.
• GitHub Codespaces Ready: Preconfigured with a devcontainer for a seamless VSCode experience in Codespaces.

Available Voices

OpenAI TTS Voices

• nova
• shimmer
• echo
• onyx
• fable
• alloy
• ash
• sage
• coral

Azure TTS Voices (Examples)

• en-US-AriaNeural
• en-US-GuyNeural
• en-US-JennyNeural
• en-GB-LibbyNeural
• en-AU-CatherineNeural
• (See Azure Voice Fonts Documentation for a full list)

Prerequisites

• Python 3.8 or higher.
• An OpenAI API key.
• An Azure Speech Service API key and region if using Azure TTS.
• GitHub Codespaces or local Docker environment supporting the devcontainer configuration.

Setup

1. Clone the Repository in GitHub Codespaces

   • Open your repository in GitHub, click the Code button, and select the Codespaces tab to create a new Codespace.

2. Configure API Keys

   • In the root directory, create a file named .env (or use Codespaces secrets) with the following content:

     OPENAI_API_KEY=your-openai-api-key
     AZURE_API_KEY=your-azure-api-key
     AZURE_REGION=your-azure-region

3. Install Dependencies

   • Open a terminal in the project root and run:

     pip install -r requirements.txt

4. Run Diagnostics (Optional)

   • Validate that your environment variables are correctly injected:

     python diagnostics/validate_env.py
     python diagnostics/validate_azure.py

5. Running the TTS Demo

   • For OpenAI TTS:

     python tts_demo.py --endpoint openai --voice-oai nova

   • For Azure TTS:

     python tts_demo.py --endpoint azure --voice en-US-AriaNeural

After running the script, check the output audio file (either output.mp3 or output_azure.mp3) using your preferred media player.

Additional Resources

• OpenAI TTS API Documentation
• Azure Speech Service Documentation
• GitHub Codespaces Documentation
• Python-dotenv Documentation

License

This project is provided for educational purposes only. Please adhere to the usage policies of OpenAI and Azure when using their services.

Acknowledgments

• Thanks to OpenAI and Azure for their TTS services.
• Special thanks to the GitHub Codespaces team for providing an excellent development environment.