README

run_cnn_chime3.sh is a Kaldi recipe script for training and decoding using a CNN-DNN + sMBR based decoder for evaluation of 
the CHiME-3 challenge dataset.
%%%%%%%
The CNN-DNN implementation is based on 
T.N. Sainath, A.-R. Mohamed, B. Kingsbury, and  B. Ramabhadran, “Deep convolutional neural networks for LVCSR,” 
in Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on, May 2013, pp. 8614–8618. 

If you use this code, please also cite
[1] D. Baby, T. Virtanen and H. Van hamme, "Coupled Dictionary-based Speech Enhancement for CHiME-3 Challenge", 
 Submitted to IEEE 2015 Automatic Speech Recognition and Understanding Workshop (ASRU), 2015.
[2] Jon Barker, Ricard Marxer, Emmanuel Vincent, and Shinji Watanabe, "The third 'CHiME' 
 Speech Separation and Recognition Challenge: Dataset, task and baselines", 
 submitted to IEEE 2015 Automatic Speech Recognition and Understanding Workshop (ASRU), 2015.

%%%%%%%
USAGE:
For using this script, copy the script to /<localpath>/CHiME3/tools/ASR/ 
Execute the run.sh script already in the CHiME-3 folder for the necessary alignment files.
If run.sh is already executed for the custom $enhancement_method, run the provided script as

run_cnn_chime3.sh $enhancement_method