[KongchangAI]
Model

ECAPA-TDNN

ECAPA-TDNN is a neural network model for speaker verification, built on the TDNN architecture, that extracts speaker embedding vectors.

Source Articles