I just open-sourced ChironX – an AI-powered guitar coaching platform that gives real-time feedback using Gemini Vision and voice feedback using Google Cloud TTS.
You upload a guitar video, and it:
- Analyzes frame-by-frame finger movement
- Detects chord transitions
- Gives personalized audio coaching (e.g., “Your index finger is too far from the fret”)
Tech stack:
- Gemini Vision API (Google)
- Google Cloud Text-to-Speech
- Flask + OpenCV + Vanilla JS
Would love open source community's feedback, ideas, and contributions. Planning to expand this into a multi-skill AI coach (piano, tennis, drawing, yoga).
You upload a guitar video, and it: - Analyzes frame-by-frame finger movement - Detects chord transitions - Gives personalized audio coaching (e.g., “Your index finger is too far from the fret”)
Tech stack: - Gemini Vision API (Google) - Google Cloud Text-to-Speech - Flask + OpenCV + Vanilla JS
GitHub: https://github.com/saurabh-yergattikar/ChironX
Would love open source community's feedback, ideas, and contributions. Planning to expand this into a multi-skill AI coach (piano, tennis, drawing, yoga).