Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement

Abstract: 

Multi-frame approaches for single-microphone speech enhancement, e.g., the multi-frame minimum-power-distortionless-response (MFMPDR) filter, are able to exploit speech correlations across neighboring time frames. In contrast to single-frame approaches such as the Wiener gain, it has been shown that multi-frame approaches achieve a substantial noise reduction with hardly any speech distortion, provided that an accurate estimate of the correlation matrices and especially the speech interframe correlation (IFC) vector is available. Typical estimation procedures of the IFC vector require an estimate of the speech presence probability (SPP) in each time-frequency (TF) bin. In this paper, we propose to use a bi-directional long short-term memory deep neural network (DNN) to estimate the SPP for each TF bin. Aiming at achieving a robust performance, the DNN is trained for various noise types and within a large signal-to-noise-ratio range. Experimental results show that the MFMPDR in combination with the proposed datadriven SPP estimator yields an increased speech quality compared to a state-of-the-art model-based SPP estimator. Furthermore, it is confirmed that exploiting interframe correlations in the MFMPDR is beneficial when compared to the Wiener gain especially in adverse scenarios.

up
0 users have voted:

Paper Details

Authors:
Marvin Tammen, Dörte Fischer, Bernd T. Meyer, Simon Doclo
Submitted On:
15 May 2020 - 6:12am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Marvin Tammen
Paper Code:
AUD-L7.3
Document Year:
2020
Cite

Document Files

ICASSP2020_Tammenetal.pdf

(16)

Subscribe

[1] Marvin Tammen, Dörte Fischer, Bernd T. Meyer, Simon Doclo, "DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5345. Accessed: Jul. 06, 2020.
@article{5345-20,
url = {http://sigport.org/5345},
author = {Marvin Tammen; Dörte Fischer; Bernd T. Meyer; Simon Doclo },
publisher = {IEEE SigPort},
title = {DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement},
year = {2020} }
TY - EJOUR
T1 - DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement
AU - Marvin Tammen; Dörte Fischer; Bernd T. Meyer; Simon Doclo
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5345
ER -
Marvin Tammen, Dörte Fischer, Bernd T. Meyer, Simon Doclo. (2020). DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement. IEEE SigPort. http://sigport.org/5345
Marvin Tammen, Dörte Fischer, Bernd T. Meyer, Simon Doclo, 2020. DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement. Available at: http://sigport.org/5345.
Marvin Tammen, Dörte Fischer, Bernd T. Meyer, Simon Doclo. (2020). "DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement." Web.
1. Marvin Tammen, Dörte Fischer, Bernd T. Meyer, Simon Doclo. DNN-Based Speech Presence Probability Estimation for Multi-Frame Single-Microphone Speech Enhancement [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5345