The project focuses on speaker recogonition techniques. The GMM model used for speaker recognition is trained on a dataset comprising of people of different age groups and gender, speaking in 3 different languages. The trained model can tested by giving a (i) Prerecorded audio file as input (ii) Real time recorded voice as input (iii) Run the entire testing sample and get accuracy of Model.The GMM model is implemented successfully with 92.857 % accuracy.
Speaker-Recogonition/ |--- Testing Audio |--- Trained_Speech_Models |--- Voice_Samples_Training |--- DataVisualization.py |--- FeatureExtraction.py |--- Testing_audio_Path.txt |--- TrainingModel.py |--- VoiceRecord.py |--- Voice_Samples_Training_Path.txt |--- main.py |--- requirements.txt |--- readme.md