-
We have taken all the files and combined them into one single file. So the model does not know how a song ends and how a song starts. It randomly picks notes and does not know that the ending of the song should be soft or the starting should be soft.
-
We are taking offset as 0.5 always. So, the implementation we have at the moment does not support varying duration of notes and different offsets between notes.
-
The instrument considered in this project is Piano. The project can be extended to more instruments like Guitar, Violin and so on.