음원 분리 기법에 대하여.. (about singing voice separation)

11개월 전

요새 제가 다니고 있는 연구실의 화두 중 하나는 음원 분리 입니다.

음원 분리는 노래처럼 여러개 악기와 목소리가 합쳐져 있는 음원을 다시 분리하는 과제 입니다.

최근 저희 연구실에서 세계 최고 수준의 성능으로 음원 분리하는 신경망 모형을 개발하였습니다.

제가 주도적으로 참여한건 아니지만 그래도 자랑해볼까 올려 봅니다. ^^;

폴킴의 허전해라는 노래에서 목소리를 분리한 데모 영상입니다. 원곡과 분리된 곡이 교차로 나옵니다. ^^;

여러가지 응용도 가능할 것 같은데.. 관심 있으신 분이 어디 계실까요...


One of the research topic in my lab is source separation.

Source separation is a task to separate the sound source from song combined with several instruments and voices.

We recently developed a CNN-based model that separates singing voice with state-of-the-art performance.

I did not take the lead but I'm going to brag about it. ^^;

This is a demo video that separates the voice from Paul Kim's song 'Empty'. The singing separated from the originals alternatively appear. ^^;

I think it's possible for various applications .. Who is interested?

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
STEEMKR.COM IS SPONSORED BY
ADVERTISEMENT
Sort Order:  trending

Congratulations @tmkor! You have completed the following achievement on the Steem blockchain and have been rewarded with new badge(s) :

You made more than 2500 comments. Your next target is to reach 3000 comments.

You can view your badges on your Steem Board and compare to others on the Steem Ranking
If you no longer want to receive notifications, reply to this comment with the word STOP

Vote for @Steemitboard as a witness to get one more award and increased upvotes!