• Publications
  • Influence
Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation
TLDR
We propose Deep Reconstruction-Classification Network (DRCN), a convolutional network that jointly learns two tasks: i) supervised source label prediction and ii) unsupervised target data reconstruction. Expand
  • 402
  • 43
  • PDF
Speech Coding and Synthesis
TLDR
An introduction to speech coding, W. Paliwal and W. Kroon a robust algorithm for pitch tracking (RAPT), D. McAulay and T. Kubin an approach to text-to-speech synthesis, R. Hedelin et al theory for transmission of vector quantization data, P. Kleijn evaluation of speech coders, J.-H. Expand
  • 604
  • 36
Domain Generalization for Object Recognition with Multi-task Autoencoders
TLDR
The problem of domain generalization is to take knowledge acquired from a number of related domains, where training data is available, and to then successfully apply it to previously unseen domains. Expand
  • 247
  • 36
  • PDF
Scatter Component Analysis: A Unified Framework for Domain Adaptation and Domain Generalization
TLDR
We propose Scatter Component Analyis (SCA), a fast representation learning algorithm that can be applied to both domain adaptation and domain generalization. Expand
  • 167
  • 34
  • PDF
Codebook driven short-term predictor parameter estimation for speech enhancement
TLDR
In this paper, we present a new technique for the estimation of short-term linear predictive parameters of speech and noise from noisy data and their subsequent use in waveform enhancement schemes. Expand
  • 182
  • 21
  • PDF
Codebook-Based Bayesian Speech Enhancement for Nonstationary Environments
TLDR
We propose a Bayesian minimum mean squared error approach for the joint estimation of the short-term predictor parameters of speech and noise, from the noisy observation. Expand
  • 136
  • 16
  • PDF
Graph-Preserving Sparse Nonnegative Matrix Factorization With Application to Facial Expression Recognition
TLDR
In this paper, a novel graph-preserving sparse nonnegative matrix factorization (GSNMF) algorithm is proposed for facial expression recognition. Expand
  • 169
  • 12
  • PDF
Entropy-constrained polar quantization and its application to audio coding
  • R. Vafin, W. Kleijn
  • Computer Science, Mathematics
  • IEEE Transactions on Speech and Audio Processing
  • 22 February 2005
TLDR
We present a new method for quantization of sinusoidal amplitudes and phases that model a short segment of an audio signal. Expand
  • 49
  • 11
Compressive sensing for sparsely excited speech signals
  • T. Sreenivas, W. Kleijn
  • Mathematics, Computer Science
  • IEEE International Conference on Acoustics…
  • 19 April 2009
TLDR
We explore a signal dependent unknown linear transform, namely the impulse response matrix operating on a sparse excitation, as in the linear model of speech production, for recovering compressed sensed speech. Expand
  • 82
  • 9
  • PDF
Quantization of LPC Parameters
Accurate reconstruction of the envelope of the short-time power spectrum is very important for both the quality and intelligibility of coded speech. For low-bit-rate speech coding, the linearExpand
  • 122
  • 9
  • PDF