For anyone unfortunate enough to stumble upon this, this is my first repository. Here, I would like to use the OpenAI Whisper model to replace the dictation function on my Mac. The reason I'm doing this is for a learning experience. I think it is beneficial for me as a young coder, and, additionally, I am very dyslexic. It is extremely difficult for me to type as quickly as I speak, and I find it much more convenient to use dictation all the time rather than typing. As such, this is why I'm here. The very rough roadmap should look something like: Experiment Whisper, Experiment MacOS, Stability.
-
Experiment with Whisper Model
- Develop a basic application to implement the Whisper model for dictation.
- Test various configurations to optimize accuracy and speed.
-
Experiment with MacOS Integration
- Address MacOS-specific challenges to integrate the dictation effectively.
- Create a seamless activation shortcut similar to existing MacOS dictation features.
-
Ensure System Stability
- Focus on the reliability and robustness of the application for daily use.
- Implement error handling and continuous performance testing.