Hi @all!
While reading the AIDA-X thread, that is hottest topic in zynthian forum by these days 
, i’ve done some research to see what else is being developed regarding neural networks and audio processing. I found this and wanted to share with you, so it don’t get lost in my busy mind and we can stay tuned about further progress:
-
A beginners introduction with some examples to listen: AI innovation helps create authentic, pitch perfect vocals | Hub
-
The implementation code, yet very experimental: GitHub - haidog-yaqub/DiffPitcher: Diffusion-based singing voice pitch correction
-
The raw academic article: https://engineering.jhu.edu/lcap/data/uploads/pdfs/waspaa2023_hai.pdf
Enjoy!