Speech enhancement based on deep denoising autoencoder

By Xugang Lu, Yu Tsao, Shigeki Matsuda, Chiori Hori. 2013

851 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Generative Models, Speech & Audio, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Generative Models

In Speech & Audio

In Signal Processing

Citations in the corpus

A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines (2018, methodology): “The authors of [142] rely on a DAE to reduce the noise and thus perform speech recognition enhancement.”
Speech Denoising with Deep Feature Losses (2018, background): “However, most pipelines still operate in the spectrogram domain [4, 5, 6, 7, 8, 9, 10, 11].”
Exploring Speech Enhancement with Generative Adversarial Networks for Robust Speech Recognition (2017, background): “networks have also been investigated in the literature for enhancement, such as denoising autoencoders [12], convolution networks [13] and recurrent networks [14].”
A Wavenet for Speech Denoising (2017, background): “Similarly, most techniques for speech denoising use magnitude spectrograms as front-end [3, 4, 5].”
SDR – Half-baked or Well Done? (2018, methodology): “Source separation and speech enhancement have been an intense focus of research in the signal processing community for several decades, and interest has gotten even stronger with the recent advent of powerful new techniques based on deep learning [1–11].”
Speech Enhancement in Multiple-Noise Conditions Using Deep Neural Networks (2016, methodology): “To directly substantiate the ability of DNN in modeling complex noisy log spectra to clean log-spectra we also measure speech distortion and noise reduction measure [14].”
Conditional Generative Adversarial Networks for Speech Enhancement and Noise-Robust Speaker Verification (2017, methodology): “Lately, deep learning methods have been used, such as DNNs [6, 8], deep autoencoders (DAEs) [5], and convolutional neural networks (CNNs) [9].”
A New Framework for CNN-Based Speech Enhancement in the Time Domain (2019, background): “popular deep learning methods are denoising autoencoders [4], feedforward neural networks [5], [6], and CNNs [7]–[9].”
Deep Learning for Environmentally Robust Speech Recognition (2017, methodology): “Specifically, in 2013, a Stacked AutoEncoder (SAE) was employed to map noisy speech to clean speech in the Mel-spectral domain (Lu et al. 2013).”
DEEP ATTRACTOR NETWORK FOR SINGLE-MICROPHONE SPEAKER SEPARATION (2016, background): “Despite the recent advances in deep learning methods for various speech processing tasks such as automatic recognition[1, 2, 3] and enhancement[4, 5, 6], speech separation remains unresolved.”
The Phasebook: Building Complex Masks via Discrete Representations for Source Separation (2019, background): “olds Rmax, phase sensitive filter (PSF: a = cos(θ) |s| |x| ), and its truncated version to [0, 1] (TPSF), ideal binary mask (IBM: a = δ(|s| > |n|)), ideal ratio mask (IRM: a = |s| / (|s|+ |n|)), and Wiener-filter-like mask (WF: a = |s| / (|s| + |n|)).”
Reverberant speech recognition combining deep neural networks and deep autoencoders augmented with a phone-class feature (2015, methodology): “Recognition of reverberant speech is performed combining “standard” DNNHMM [9] decoding and a feature enhancement through deep autoencoder [10][11].”
Raw waveform-based speech enhancement by fully convolutional networks (2017, background): “Recently, deep denoising autoencoder (DDAE) and deep neural network (DNN)-based SE models have also been proposed and extensively investigated [11-13].”
Cycle-Consistent Speech Enhancement (2018, methodology): “To deal with this problem, the feature mapping approach [12, 13, 14, 15, 16, 17] was proposed to train a mapping network that directly transforms the noisy features to”