GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → Doyosae → Phase-aware-Deep-Complex-UNet

Doyosae / Phase-aware-Deep-Complex-UNet

Licence: other
(NOT Official) Implementation DC-UNet (ICLR 2019)

Programming Languages

python
139335 projects - #7 most used programming language

Labels

deep-learningcomplex-networksspeech-enhancement

Projects that are alternatives of or similar to Phase-aware-Deep-Complex-UNet

simplagion
Code associated to the paper 'Simplicial models of social contagion' Nat. Commun. (2019)
Stars: ✭ 19 (-60.42%)
Mutual labels:  complex-networks
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 d…
Stars: ✭ 25 (-47.92%)
Mutual labels:  speech-enhancement
brainGraph
Graph theory analysis of brain MRI data
Stars: ✭ 136 (+183.33%)
Mutual labels:  complex-networks
sorn
PyPi Package of Self-Organizing Recurrent Neural Networks (SORN) and Neuro-robotics using OpenAI Gym
Stars: ✭ 16 (-66.67%)
Mutual labels:  complex-networks
NonlinearDynamicsTextbook
Nonlinear Dynamics: A concise introduction interlaced with code
Stars: ✭ 145 (+202.08%)
Mutual labels:  complex-networks
SpleeterRT
Real time monaural source separation base on fully convolutional neural network operates on Time-frequency domain.
Stars: ✭ 111 (+131.25%)
Mutual labels:  speech-enhancement
DEMON
DEMON: a local-first discovery method for overlapping communities.
Stars: ✭ 26 (-45.83%)
Mutual labels:  complex-networks
fdndlp
A speech dereverberation algorithm, also called wpe
Stars: ✭ 115 (+139.58%)
Mutual labels:  speech-enhancement
torchsubband
Pytorch implementation of subband decomposition
Stars: ✭ 63 (+31.25%)
Mutual labels:  speech-enhancement
football-graphs
Graphs and passing networks in football.
Stars: ✭ 81 (+68.75%)
Mutual labels:  complex-networks
dynetx
Dynamic Network Analysis library
Stars: ✭ 75 (+56.25%)
Mutual labels:  complex-networks
Awesome Network Analysis
A curated list of awesome network analysis resources.
Stars: ✭ 2,525 (+5160.42%)
Mutual labels:  complex-networks
Speech Enhancement MMSE-STSA
A statistical model-based Speech Enhancement Using MMSE-STSA
Stars: ✭ 54 (+12.5%)
Mutual labels:  speech-enhancement
time series clustering via community detection
Code used in the paper "Time Series Clustering via Community Detection in Networks"
Stars: ✭ 27 (-43.75%)
Mutual labels:  complex-networks
CommunityDetection
复杂网络社区发现算法、评价指标、常用数据集整理
Stars: ✭ 171 (+256.25%)
Mutual labels:  complex-networks
noisy-networks-measurements
Noisy network measurement with stan
Stars: ✭ 42 (-12.5%)
Mutual labels:  complex-networks
deepbeam
Deep learning based Speech Beamforming
Stars: ✭ 58 (+20.83%)
Mutual labels:  speech-enhancement
ComplexCi
The c++ implementation of Collective Influence (CI) algorithm in Complex Network with DataCastle Competition Solution
Stars: ✭ 30 (-37.5%)
Mutual labels:  complex-networks
awesome-speech-enhancement
A curated list of awesome Speech Enhancement papers, libraries, datasets, and other resources.
Stars: ✭ 48 (+0%)
Mutual labels:  speech-enhancement
speech-enhancement-WGAN
speech enhancement GAN on waveform/log-power-spectrum data using Improved WGAN
Stars: ✭ 35 (-27.08%)
Mutual labels:  speech-enhancement
View All Similar Projects ➔

Introuduction

Impelmentation Phase-aware Speech Enhnacement Deep Complex UNet
This is convolution neural networks model for Speech Enhancement
Papers URL

  1. Phase-aware Speech Enhancement Deep Complex UNet - openreview
  2. Phase-aware Speech Enhancement Deep Complex UNet - arxiv

Architecture

archi

Issue!

Don't use DCUnet16 and DCUnet20 via ComplexBatchNormalization
You can use only Naive_DCUnet series
컴플렉스 배치 정규화를 적용한 DCUnet16, DCUnet20을 사용하지 마세요.
(텐서플로우 최신 버전에서는 위 레이어를 위한 기능이 삭제되었습니다.)
Naive하게 배치 정규화를 사용한 모델만 참고하세요.

Requirements

Python >= 3.6.9 (3.7.6)
numpy
scipy
librosa == 0.7.2 (with numba 0.48.0)
Tensorflow == 2.1.0

Folder

Directory
./Deep_Complex_UNet
    /datasets
        /train_noisy
        /train_clean
        /test_noisy
        /test_clean
    /model_save
    /model_pred
    /complex_layers
        __init__.py
        stft.py
        layer.py
        activation.py
        normaliztion.py
        dcunet.py
    model_module.py
    model_loss.py
    model_data.py
    model_test.py
    model_train.py

Usage

--model naive_dcunet16 naive_dcunet20 dcunet16 dcunet20
--loss SDR or wSDR  
--save "save file name"  
python model_train.py --model dcunet20 --loss wSDR --save dcunet20_default --optim adam --lr 0.002
python model_train.py --model dcunet16 --loss wSDR --save dcunet16_default --optim adam --lr 0.002
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].