I\'m working on a speaker recognizer project on TensorFlow, to train my model I chose Voxceleb data set, the problem is that I don\'t know how to load my training and testin