how to run.txt

DNN based speech enhancement tool (developed by Dr. yong xu)

Note: Matlab should be installed in your PC first. Perl should be installed in your PC.
This zip contains the best DNN model trained by us.
This exe can be used to enhance any noisy speech, even from real world noisy speech!

Enhancing steps:

1 step0: clear all of the files in "wav_lsp", and then copy all of your noisy waveforms into "wav_lsp"

2 step1: in Matlab, JUST RUN "step1_DNNenh_for16kHz.m" without any parameters to be tuned and then all of the noisy waveforms will be obtained in "wav_lsp"
(There is a noisy 16kHz waveform in wav_lsp/timit_noisy_SNR5.wav  as an example)

or you can just click: step1_DNNenh_for16kHz.exe, everything will be done for you!




all is done!!!

The related enhanced demos:
http://home.ustc.edu.cn/~xuyong62/demo/IS15.html
http://home.ustc.edu.cn/~xuyong62/demo/SE_DNN_taslp.html

If you have any questions, please feel free to contact Dr. Yong XU (yong.xu.ustc@gmail.com)

Please cite:
[1] Multi-objective learning and Mask-based Post-processing for Deep Neural Network based Speech Enhancement.
Yong Xu, Jun Du, Zhen Huang, Li-Rong Dai, Chin-Hui Lee, accepted, Interspeech2015, Dresden, Germany
[2] A Regression Approach to Speech Enhancement Based on Deep Neural Networks.
Yong Xu, Jun Du, Li-Rong Dai and Chin-Hui Lee, accepted to IEEE/ACM Transactions on Audio, Speech, and Language Processing, P. 7-19, Vol.
23, No. 1, IEEE/ACM trans. on audio, speech, language processing, 2015
[3] An Experimental Study on Speech Enhancement Based on Deep Neural Networks.
Yong Xu, Jun Du, Li-Rong Dai and Chin-Hui Lee, IEEE signal processing letters, p. 65-68, vol. 21, no. 1, January 2014
[4] SPEECH SEPARATION METHOD AND SYSTEM, US patent, US 20160189730A1

Note: if you want to use it for commercial product, please contact Dr. Yong XU (yong.xu.ustc@gmail.com)