Picture for Mads Græsbøll Christensen

Mads Græsbøll Christensen

Provable Privacy Advantages of Decentralized Federated Learning via Distributed Optimization

Add code
Jul 12, 2024
Viaarxiv icon

Frequency bin-wise single channel speech presence probability estimation using multiple DNNs

Add code
Feb 23, 2023
Figure 1 for Frequency bin-wise single channel speech presence probability estimation using multiple DNNs
Figure 2 for Frequency bin-wise single channel speech presence probability estimation using multiple DNNs
Figure 3 for Frequency bin-wise single channel speech presence probability estimation using multiple DNNs
Figure 4 for Frequency bin-wise single channel speech presence probability estimation using multiple DNNs
Viaarxiv icon

A Two-Stage Deep Representation Learning-Based Speech Enhancement Method Using Variational Autoencoder and Adversarial Training

Add code
Nov 16, 2022
Figure 1 for A Two-Stage Deep Representation Learning-Based Speech Enhancement Method Using Variational Autoencoder and Adversarial Training
Figure 2 for A Two-Stage Deep Representation Learning-Based Speech Enhancement Method Using Variational Autoencoder and Adversarial Training
Figure 3 for A Two-Stage Deep Representation Learning-Based Speech Enhancement Method Using Variational Autoencoder and Adversarial Training
Figure 4 for A Two-Stage Deep Representation Learning-Based Speech Enhancement Method Using Variational Autoencoder and Adversarial Training
Viaarxiv icon

Privacy-Preserving Distributed Expectation Maximization for Gaussian Mixture Model using Subspace Perturbation

Add code
Sep 16, 2022
Figure 1 for Privacy-Preserving Distributed Expectation Maximization for Gaussian Mixture Model using Subspace Perturbation
Figure 2 for Privacy-Preserving Distributed Expectation Maximization for Gaussian Mixture Model using Subspace Perturbation
Figure 3 for Privacy-Preserving Distributed Expectation Maximization for Gaussian Mixture Model using Subspace Perturbation
Viaarxiv icon

A deep representation learning speech enhancement method using $β$-VAE

Add code
May 11, 2022
Figure 1 for A deep representation learning speech enhancement method using $β$-VAE
Figure 2 for A deep representation learning speech enhancement method using $β$-VAE
Figure 3 for A deep representation learning speech enhancement method using $β$-VAE
Figure 4 for A deep representation learning speech enhancement method using $β$-VAE
Viaarxiv icon

A Bayesian Permutation training deep representation learning method for speech enhancement with variational autoencoder

Add code
Jan 24, 2022
Figure 1 for A Bayesian Permutation training deep representation learning method for speech enhancement with variational autoencoder
Figure 2 for A Bayesian Permutation training deep representation learning method for speech enhancement with variational autoencoder
Figure 3 for A Bayesian Permutation training deep representation learning method for speech enhancement with variational autoencoder
Figure 4 for A Bayesian Permutation training deep representation learning method for speech enhancement with variational autoencoder
Viaarxiv icon

Communication efficient privacy-preserving distributed optimization using adaptive differential quantization

Add code
May 30, 2021
Figure 1 for Communication efficient privacy-preserving distributed optimization using adaptive differential quantization
Figure 2 for Communication efficient privacy-preserving distributed optimization using adaptive differential quantization
Figure 3 for Communication efficient privacy-preserving distributed optimization using adaptive differential quantization
Figure 4 for Communication efficient privacy-preserving distributed optimization using adaptive differential quantization
Viaarxiv icon

Speech Decomposition Based on a Hybrid Speech Model and Optimal Segmentation

Add code
May 04, 2021
Figure 1 for Speech Decomposition Based on a Hybrid Speech Model and Optimal Segmentation
Figure 2 for Speech Decomposition Based on a Hybrid Speech Model and Optimal Segmentation
Figure 3 for Speech Decomposition Based on a Hybrid Speech Model and Optimal Segmentation
Viaarxiv icon

A Variational EM Method for Pole-Zero Modeling of Speech with Mixed Block Sparse and Gaussian Excitation

Add code
Jun 24, 2017
Figure 1 for A Variational EM Method for Pole-Zero Modeling of Speech with Mixed Block Sparse and Gaussian Excitation
Figure 2 for A Variational EM Method for Pole-Zero Modeling of Speech with Mixed Block Sparse and Gaussian Excitation
Figure 3 for A Variational EM Method for Pole-Zero Modeling of Speech with Mixed Block Sparse and Gaussian Excitation
Figure 4 for A Variational EM Method for Pole-Zero Modeling of Speech with Mixed Block Sparse and Gaussian Excitation
Viaarxiv icon