Instruction file imported from donghaozhang/gemini-tts-multi-speaker (.cursor/rules/development-workflow.mdc). Copyright stays with the author.

Development Workflow

Environment Setup

1. API Key Configuration: Copy .env.example to .env and add your Gemini API key
2. Dependencies: Install with pip install -r requirements.txt from requirements.txt
3. Testing: Use batch files for quick testing of different scenarios

File Organization

• Source Code: All Python files in project root
• Generated Audio: All WAV files automatically saved to result/ directory
• Documentation: README.md and PROJECT_STRUCTURE.md

Git Workflow

The .gitignore excludes:

• Environment files (.env)
• Audio files (*.wav, *.mp3) - too large for version control
• Python cache (__pycache__/)
• IDE files and temporary files

Adding New Scenarios

To add new conversation scenarios:

1. Single Speaker: Add functions to gemini_tts_example.py
2. Multi-Speaker: Add demo functions to multi_speaker_demo.py
3. Academic Content: Add presentations to academic_papers_demo.py

Batch Script Pattern

All batch files follow this pattern:

@echo off
echo Loading environment variables from .env file...
for /f "delims=" %%x in (.env) do (set "%%x")
echo Running [Description]...
python [script_name].py
pause

Testing Strategy

Use run_demos.bat to launch any of the demos or run them all at once.

Code Style

• Functions use descriptive names with clear purposes
• Error handling includes user-friendly messages
• Voice names are converted to lowercase for API compatibility
• Output files are saved with descriptive names in result/ directory

Extension Points

• Add new voice combinations in speaker arrays
• Create new dialogue scenarios with different themes
• Implement additional audio processing features
• Add support for different output formats