Hui-1/speaker_recognition-campplus-zh-en-16k-base

CAM++: A Fast and Efficient Speaker Recognition Network

CAM++ is a fast and efficient network based on a densely-connected time delay neural network (D-TDNN). This repository provides tools for extracting speaker embeddings and performing speaker verification tasks with a pretrained CAM++ model. It has been trained on a large-scale training dataset, which includes Chinese and English corpora.