0.3

• README now notes that this depends on ffmpeg, can't be run in the same environment as LLM, and currently supports Python 3.11 but not Python 3.12.
• Improvements to audio quality when exporting as MP3. Thanks, Mikolaj Holysz. #7
• New -m/--model option for switching the model from the default tts-1 to tts-1-hd or other models. Thanks, Mikolaj Holysz. #5