We want to be able to listen to a song without vocals.

What it does

Separates vocals from music for any song that you want

How we built it

We used web scraping to get the song from youtube. Then, we used librosa to get all the notes and turn it into a mel spectrogram graph. After that, we would get the machine/AI to extract all the lyrics out of the mel spectrogram graph. Then it would download the version of the song you picked that doesn't have vocals.

Challenges we ran into

We did not have enough time to fully complete the project and we ran into some bugs.

Accomplishments that we're proud of

We successfully turned music into a mel spectrogram graph.

What we learned

A project like this would take more than a few days to complete. There are a lot of complications that we didn't expect when we started the project.

What's next for vocal separator

To send it out to the world so people can enjoy music without vocals if they like.

Built With

  • lab
  • librosa
  • math
  • python
  • spectrogram
Share this project: