The goal is to train a machine learning model from a custom dataset of water pouring sounds. The final system will generate sound triggers based on the pouring speed, potentially creating an orchestral experience in a collaborative web app where multiple users can interact simultaneously.
- The idea is to create our own database of the sounds.
- We need to use Pytorch with NN to create our model.
- The goal is having some values in the and to use as controller or trigger of music.
To follow the process, you can see the notebook Let_it_flow_rra that tells more or less the steps that we followed.
