Updated to fulfill dependency for executing audio processing in Sagemaker environment.
New Demo Video: https://youtu.be/h_ugT3hYzH8
New Demo Instructions(demo is now simplified):
- Clone this repository as a notebook instance in AWS Sagemaker.
- Go through main.ipynb file in the repository. Execute all the code blocks.
- The last code block of the notebook allows you to listen to the generated song.
Version 2.1 of AI Cover Musician now uses Tensorflow model to detect pitch in the generated audio and the isolated vocals. This allows for pitch modification to try and match the original song.