Project presenting usage of the GMM method for feature extraction from sepeech signals.
Presented method was compared to MFCC - the the most popular method used currently for this purpose.
Data
- directory containing zipped training and test samplesApplication_of_Gaussian_Mixture_Models_for_fe.pdf
- fake paper, presenting theroretical aspects of speech signals, MFCC and GMM methodsproject.ipynb
- notebook containing project coderequirements.txt
- requirements file used for setting up python environment
All other
.py
contains debug code or code fragments extracted from.ipynb
for unknown reason :)