Experiments on Open-Set Speaker Identification with Discriminatively Trained Neural Networks

04/02/2019

∙

This paper presents a study on discriminative artificial neural network classifiers in the context of open-set speaker identification. Both 2-class and multi-class architectures are tested against the conventional Gaussian mixture model based classifier on enrolled speaker sets of different sizes. The performance evaluation shows that the multi-class neural network system has superior performance for large population sizes.

READ FULL TEXT

Experiments on Open-Set Speaker Identification with Discriminatively Trained Neural Networks

Novel Cascaded Gaussian Mixture Model-Deep Neural Network Classifier for Speaker Identification in Emotional Talking Environments

Optimizing Multi-Taper Features for Deep Speaker Verification

Latent space representation for multi-target speaker detection and identification with a sparse dataset using Triplet neural networks

CASA-Based Speaker Identification Using Cascaded GMM-CNN Classifier in Noisy and Emotional Talking Conditions

Single neuron-based neural networks are as efficient as dense deep neural networks in binary and multi-class recognition problems

Histogram Transform-based Speaker Identification

Computing Class Hierarchies from Classifiers

Experiments on Open-Set Speaker Identification with Discriminatively Trained Neural Networks

Related Research

Novel Cascaded Gaussian Mixture Model-Deep Neural Network Classifier for Speaker Identification in Emotional Talking Environments

Optimizing Multi-Taper Features for Deep Speaker Verification

Latent space representation for multi-target speaker detection and identification with a sparse dataset using Triplet neural networks

CASA-Based Speaker Identification Using Cascaded GMM-CNN Classifier in Noisy and Emotional Talking Conditions

Single neuron-based neural networks are as efficient as dense deep neural networks in binary and multi-class recognition problems

Histogram Transform-based Speaker Identification

Computing Class Hierarchies from Classifiers