Add FastAPI-based OpenAI-compatible Text-to-Speech API and Audio Format Conversion Tools
and clear ipynb output to shrink its size