Speaker verification is the verifying the identity of a person from characteristics of the voice.
We propose the use of a coupled 3D Convolutional Neural Network (3D-CNN) architecture that can map both modalities into a representation space to evaluate the correspondence of audio-visual streams using the learned multimodal features.
In our paper, we propose an adaptive feature learning by utilizing the 3D-CNNs for direct speaker model creation in which, for both development and enrollment phases, an identical number of spoken utterances per speaker is fed to the network for representing the speakers' utterances and creation of the speaker model.
Rather than employing standard hand-crafted features, the latter CNNs learn low-level speech representations from waveforms, potentially allowing the network to better capture important narrow-band speaker characteristics such as pitch and formants.
In this paper, we propose a new loss function called generalized end-to-end (GE2E) loss, which makes the training of speaker verification models more efficient than our previous tuple-based end-to-end (TE2E) loss function.
In this paper we present a data-driven, integrated approach to speaker verification, which maps a test utterance and a few reference utterances directly to a single score for verification and jointly optimizes the system's components using the same evaluation protocol and metric as at test time.
We present a factorized hierarchical variational autoencoder, which learns disentangled and interpretable representations from sequential data without supervision.
Deep generative models have achieved great success in unsupervised learning with the ability to capture complex nonlinear relationships between latent generating factors and observations.
In this work, we propose our replay attacks detection system - Attentive Filtering Network, which is composed of an attention-based filtering mechanism that enhances feature representations in both the frequency and time domains, and a ResNet-based classifier.