SuperLectures.com

MODEL-BASED SPEECH ENHANCEMENT USING SNR DEPENDENT MMSE ESTIMATION

Full Paper at IEEE Xplore

Speech Enhancement

Presented by: Thomas Esch , Author(s): Thomas Esch, Peter Vary, RWTH Aachen University, Germany

This contribution presents a modified Kalman filter approach for single channel speech enhancement which is operating in the frequency domain. In the first step, temporal correlation of successive frames is exploited yielding estimates of the current speech and noise DFT coefficients. This first prediction is updated in the second step applying an SNR dependent MMSE estimator which is adapted to the (measured) statistics of the speech prediction error signal. Objective measurements show consistent improvements compared to estimators which do not take into account the temporal correlation or the influence of the input SNR on the statistics of the prediction error signal.

  Speech Transcript

|

  Slides

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

0:00:16

  1. slide

0:01:00

  2. slide

0:01:42

  3. slide

0:03:26

  4. slide

0:04:31

  5. slide

0:05:11

  6. slide

0:06:20

  7. slide

0:07:26

  8. slide

0:08:13

  9. slide

0:09:25

 10. slide

0:09:51

 11. slide

0:13:04

 12. slide

0:13:56

 13. slide

0:14:33

 14. slide

  Comments

Please sign in to post your comment!

  Lecture Information

Recorded: 2011-05-27 17:55 - 18:15, Panorama
Added: 9. 6. 2011 05:05
Number of views: 122
Video resolution: 1024x576 px, 512x288 px
Video length: 0:16:47
Audio track: MP3 [5.73 MB], 0:16:47