SuperLectures.com

SCORE INFORMED AUDIO SOURCE SEPARATION USING A PARAMETRIC MODEL OF NON-NEGATIVE SPECTROGRAM

Full Paper at IEEE Xplore

Music Signal Processing

Presented by: Romain Hennequin, Author(s): Romain Hennequin, Bertrand David, Roland Badeau, Institut TELECOM / TELECOM ParisTech, France

In this paper we present a new technique for monaural source separation in musical mixtures, which uses the knowledge of the musical score. This information is used to initialize an algorithm which computes a parametric decomposition of the spectrogram based on non-negative matrix factorization (NMF). This algorithm provides time-frequency masks which are used to separate the sources with Wiener filtering.

  Speech Transcript

|

  Slides

Enlarge the slide | Show all slides in a pop-up window

0:00:16

  1. slide

0:00:38

  2. slide

0:00:55

  3. slide

0:01:50

  4. slide

0:02:51

  5. slide

0:04:00

  6. slide

0:04:29

  7. slide

0:05:57

  8. slide

0:06:32

  9. slide

0:07:31

 10. slide

0:08:53

 11. slide

0:09:35

 12. slide

0:10:10

 13. slide

0:11:48

 14. slide

0:12:10

 15. slide

0:13:19

 16. slide

0:13:55

 17. slide

0:14:40

 18. slide

0:14:53

 19. slide

0:14:59

 20. slide

0:16:02

 21. slide

0:16:51

 22. slide

0:18:05

 23. slide

0:19:56

 24. slide

0:21:18

 25. slide

  Comments

Please sign in to post your comment!

  Lecture Information

Recorded: 2011-05-24 15:25 - 15:45, Club H
Added: 9. 6. 2011 00:19
Number of views: 23
Video resolution: 1024x576 px, 512x288 px
Video length: 0:23:49
Audio track: MP3 [8.15 MB], 0:23:49