fdeng1983 Goto Github PK

abstention

Abstention Library: Methods for Computing Abstention Scores

audio-visual-speech-enhancement

Official Implementation of "Visual Speech Enhancement", Interspeech 2018.

awesome-ai-best-papers

A curated list of famous CV/NLP/ML/AI best papers, inspired by Best Paper Awards in Computer Science (since 1996).

clearwave

Denoise Speech (Enhanced Speech or Speech enhancement) by Deep Learning (Using Keras and Tensorflow)

clustering-with-deep-learning

Generic implementation for clustering with deep learning : representation learning (DNN) + clustering

cnn-for-single-channel-speech-enhancement

Convolutional neural nets for single channel speech enhancement

cnn-speech-denoising

cnn_finetune

Fine-tune CNN in Keras

collaborative-denoising-auto-encoder

Collaborative Denoising Auto-Encoder for Top-N Recommender Systems (CDAE)

compare_gan

Compare GAN code.

convolutionaneuralnetworkstoenhancecodedspeech

In this work we propose two postprocessing approaches applying convolutional neural networks (CNNs) either in the time domain or the cepstral domain to enhance the coded speech without any modification of the codecs. The time domain approach follows an end-to-end fashion, while the cepstral domain approach uses analysis-synthesis with cepstral domain features. The proposed postprocessors in both domains are evaluated for various narrowband and wideband speech codecs in a wide range of conditions. The proposed postprocessor improves speech quality (PESQ) by up to 0.25 MOS-LQO points for G.711, 0.30 points for G.726, 0.82 points for G.722, and 0.26 points for adaptive multirate wideband codec (AMR-WB). In a subjective CCR listening test, the proposed postprocessor on G.711-coded speech exceeds the speech quality of an ITU-T-standardized postfilter by 0.36 CMOS points, and obtains a clear preference of 1.77 CMOS points compared to G.711, even en par with uncoded speech.

ddrnet

Depth Map Denoising and Refinement for Consumer Depth Cameras Using Cascaded CNNs (ECCV 2018)

deep-clustering

deep clustering method for single-channel speech separation

deep-voice-conversion

Deep neural networks for voice conversion (voice style transfer) in Tensorflow

deepxi

Deep Xi: A Deep Learning Approach to A Priori SNR Estimation. Used for Speech Enhancement and robust ASR.

demucs

Code for the paper Music Source Separation in the Waveform Domain

densenet

Densely Connected Convolutional Networks, In CVPR 2017 (Best Paper Award).

dncnn

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising (TIP, 2017)

dncnn-tensorflow

:octocat::octocat:A tensorflow implement of the paper "Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising"

feed-forward-style-transfer

TensorFlow implementation of real-time style transfer using feed-forward generation. This builds on the original style-transfer algorithm and allows for common personal computers to transform images.

ffdnet

FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoising (TIP, 2018)

fftnet

A PyTorch implementation of the FFTNet: a Real-Time Speaker-Dependent Neural Vocoder

g2p-seq2seq

G2P with Tensorflow

generative-models

Collection of generative models, e.g. GAN, VAE in Pytorch and Tensorflow.

keras

Deep Learning for humans

keras-gan

Keras implementations of Generative Adversarial Networks.

keras-tcn

Keras Temporal Convolutional Network.

lstm_pit_speech_separation

Multi-speaker Speech Separation with LSTM/BLSTM by Permutation Invariant Training method.

magenta

Magenta: Music and Art Generation with Machine Intelligence

mcse

Multi-channel speech enhancement system (MVDR beamformer + several postfilters)