• Publications
  • Influence
A Regression Approach to Speech Enhancement Based on Deep Neural Networks
TLDR
In contrast to the conventional minimum mean square error (MMSE)-based noise reduction techniques, we propose a supervised method to enhance speech by means of finding a mapping function between noisy and clean speech signals based on deep neural networks (DNNs). Expand
  • 717
  • 71
  • PDF
An Experimental Study on Speech Enhancement Based on Deep Neural Networks
TLDR
This letter presents a regression-based speech enhancement framework using deep neural networks (DNNs) with multiple-layer deep architecture. Expand
  • 584
  • 37
  • PDF
The Fixed-Size Ordinally-Forgetting Encoding Method for Neural Network Language Models
TLDR
In this paper, we propose the new fixedsize ordinally-forgetting encoding (FOFE) method, which can almost uniquely encode any variable-length sequence of words into a fixed-size representation. Expand
  • 63
  • 16
  • PDF
WaveNet Vocoder with Limited Training Data for Voice Conversion
TLDR
This paper investigates the approaches of building WaveNet vocoder with limited training data for voice conversion (VC). Expand
  • 89
  • 15
  • PDF
Feedforward Sequential Memory Networks: A New Structure to Learn Long-term Dependency
TLDR
We propose a novel neural network structure, namely \emph{feedforward sequential memory networks (FSMN)}, to model long-term dependency in time series without using recurrent feedback. Expand
  • 52
  • 13
  • PDF
Dynamic noise aware training for speech enhancement based on deep neural networks
TLDR
We propose three algorithms to address the mismatch problem in deep neural network (DNN) based speech enhancement. Expand
  • 68
  • 11
  • PDF
Multiple-target deep learning for LSTM-RNN based speech enhancement
TLDR
In this study, we explore long short-term memory recurrent neural networks (LSTM-RNNs) for speech enhancement. Expand
  • 85
  • 9
  • PDF
Robust speech recognition with speech enhanced deep neural networks
TLDR
We propose a signal pre-processing front-end to enhance speech based on deep neural networks (DNNs) and use the enhanced speech features directly to train hidden Markov models (HMMs) for robust speech recognition. Expand
  • 93
  • 8
  • PDF
Voice Conversion Using Deep Neural Networks With Layer-Wise Generative Training
TLDR
This paper presents a new spectral envelope conversion method using deep neural networks (DNNs) to construct a global non-linear mapping relationship between the spectral envelopes of two speakers. Expand
  • 181
  • 7
Fast Adaptation of Deep Neural Network Based on Discriminant Codes for Speech Recognition
TLDR
Fast adaptation of deep neural networks (DNN) is an important research topic in deep learning. Expand
  • 133
  • 7