The UPC Speaker Verification System Submitted to VoxCeleb Speaker Recognition Challenge 2020 (VoxSRC-20)
This is an incremental improvement for speaker recognition in audio processing, targeting the VoxSRC-20 challenge.
The authors tackled speaker verification by developing three systems combining autoencoder and CNN-based siamese architectures, achieving results on VoxCeleb-1 and VoxSRC-20 datasets.
This report describes the submission from Technical University of Catalonia (UPC) to the VoxCeleb Speaker Recognition Challenge (VoxSRC-20) at Interspeech 2020. The final submission is a combination of three systems. System-1 is an autoencoder based approach which tries to reconstruct similar i-vectors, whereas System-2 and -3 are Convolutional Neural Network (CNN) based siamese architectures. The siamese networks have two and three branches, respectively, where each branch is a CNN encoder. The double-branch siamese performs binary classification using cross entropy loss during training. Whereas, our triple-branch siamese is trained to learn speaker embeddings using triplet loss. We provide results of our systems on VoxCeleb-1 test, VoxSRC-20 validation and test sets.