-
Notifications
You must be signed in to change notification settings - Fork 110
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Style interpolation generation #36
Comments
I think this will be difficult to train, but you can find composers between two styles like Beethoven's music style is a transition from Baroque to Romatic styles. |
It's a generation-level feature. You don't have to train it on interpolation. We have a paper written on the details that we may release in the future. |
@calclavia it's not as simple as mixture of two styles of melodies. xD |
I think the vertical structure is difficult,for example, it's difficult to mix polyphonic music and homophonic music. |
Allow the generation algorithm to interpolate between two styles.
Given two styles as vectors, we want the algorithm to slowly change style every single time step, until it ends up with a different style by the end of the song.
E.g:
[1, 0, 0]
->[0.5, 0.5, 0]
->[0, 1, 0]
This should work with a mixture of styles as well
[0.5, 0.5, 0]
to[0, 0.5, 0.5]
The text was updated successfully, but these errors were encountered: