Trained speaker embedding deep learning models and evaluation pipelines in pytorch and tesorflow for speaker recognition.