Skip to content
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

Amazing job!! This opens the door for new research & possibilities #5

Open
ParasharaRamesh opened this issue Aug 30, 2021 · 1 comment

Comments

@ParasharaRamesh
Copy link

This is a really amazing effort and an outstanding achievement in my opinion.

Here are my thoughts on future enhancements which if solved can open up opportunities for carnatic music generation by deep learning/machine learning models

This is definitely an arduous task, but if we somehow can reach a stage where we:

  1. Have search apis for searching across composers/songs/ragas/talas in a programmatic manner
  2. Have every second of every video present, notated in a uniform manner that not only captures all gamakas effectively but also can be rendered in a truly carnatic fashion by some piece of software
  3. Integrate with youtube apis and other apis for data cleanup such that all songs are normalized in the same shruti, redundant seconds are removed, background noise is cleaned up etc

Then this would truly function as the first carnatic music dataset which can be used for performing research a wide array of exciting problems such as:

  1. Raga alapana generation
  2. Raga identifier
  3. Raga notater
  4. Tala identifier

So on and so forth.

Similar kinds of problems have already been solved in the generation of western classical music which is a slightly easier problem to solve because of the lack of gamakas and the presense of midi notation.

Its best if we can get a conversation started around this so that we can solve this problem as a community

Hope we reach such a stage one day so that we can truly unlock the potential of indian classical music :)

@jeevster
Copy link

Hello! I trained a large-scale AI model on data from this collection to detect Carnatic ragas. It is capable of identifying 150 ragams.

https://huggingface.co/spaces/jeevster/carnatic-raga-classifier

Agreed that this is an invaluable collection, and the future possibilities are endless!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants