Voice recognition based on CNN

• The folder "data" contains all data ( 3 persons with 200 recordings ).
• The folder "sounds" contains some voices that don't exist in the training data.
• The file "preprocess.py" contains the functions to extract mfcc features and to split data.
• The file "RecVocal.ipynb" and "RecVocal000.ipynb" contains the different CNN models with the final accurancy and related results.
• The file "Rapport_Final_PPP_Amal_Hajer.pdf" is a report for all tests and results.