Feature extraction pipeline for various commonly used speech features
###############
Usage:
Job filelist- This is the list of wav files for which you want features extracted and
the destination paths for the extracted features
Format of the filelist:
<Full/Path/to/wav/file>,<Path/to/output/file>
config file - Look at the example_config.conf for the sample config file
The config file specifies the feature type, window length, frame shift etc. for feature extraction
Running the code: python extractFeatures.py <job_filelist> <config_file>
Example: python extractFeatures.py example_flist.txt example_config.conf
############### Output format The extracted features are saved in the HTK format.