All Projects → leftthomas → Segan

leftthomas / Segan

A PyTorch implementation of SEGAN based on INTERSPEECH 2017 paper "SEGAN: Speech Enhancement Generative Adversarial Network"

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Segan

Iseebetter
iSeeBetter: Spatio-Temporal Video Super Resolution using Recurrent-Generative Back-Projection Networks | Python3 | PyTorch | GANs | CNNs | ResNets | RNNs | Published in Springer Journal of Computational Visual Media, September 2020, Tsinghua University Press
Stars: ✭ 202 (+146.34%)
Mutual labels:  cnn, gan
Pytorch Srgan
A modern PyTorch implementation of SRGAN
Stars: ✭ 289 (+252.44%)
Mutual labels:  cnn, gan
Srgan
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
Stars: ✭ 2,641 (+3120.73%)
Mutual labels:  cnn, gan
Tensorflow Tutorials
텐서플로우를 기초부터 응용까지 단계별로 연습할 수 있는 소스 코드를 제공합니다
Stars: ✭ 2,096 (+2456.1%)
Mutual labels:  cnn, gan
Tensorflow Tutorial
Tensorflow tutorial from basic to hard, 莫烦Python 中文AI教学
Stars: ✭ 4,122 (+4926.83%)
Mutual labels:  cnn, gan
Keraspp
코딩셰프의 3분 딥러닝, 케라스맛
Stars: ✭ 178 (+117.07%)
Mutual labels:  cnn, gan
Mydeeplearning
A deep learning library to provide algs in pure Numpy or Tensorflow.
Stars: ✭ 281 (+242.68%)
Mutual labels:  cnn, gan
DLSS
Deep Learning Super Sampling with Deep Convolutional Generative Adversarial Networks.
Stars: ✭ 88 (+7.32%)
Mutual labels:  cnn, gan
Time Series Prediction
A collection of time series prediction methods: rnn, seq2seq, cnn, wavenet, transformer, unet, n-beats, gan, kalman-filter
Stars: ✭ 351 (+328.05%)
Mutual labels:  cnn, gan
Basicocr
BasicOCR是一个致力于解决自然场景文字识别算法研究的项目。该项目由长城数字大数据应用技术研究院佟派AI团队发起和维护。
Stars: ✭ 336 (+309.76%)
Mutual labels:  cnn, gan
Anime Face Gan Keras
A DCGAN to generate anime faces using custom mined dataset
Stars: ✭ 161 (+96.34%)
Mutual labels:  cnn, gan
Vc With Gan
Voice Conversion with GANs
Stars: ✭ 13 (-84.15%)
Mutual labels:  gan, voice
Visual Feature Attribution Using Wasserstein Gans Pytorch
Implementation of Visual Feature Attribution using Wasserstein GANs (VAGANs, https://arxiv.org/abs/1711.08998) in PyTorch
Stars: ✭ 88 (+7.32%)
Mutual labels:  cnn, gan
Cyclegan Music Style Transfer
Symbolic Music Genre Transfer with CycleGAN
Stars: ✭ 201 (+145.12%)
Mutual labels:  cnn, gan
Dcpdn
Densely Connected Pyramid Dehazing Network (CVPR'2018)
Stars: ✭ 321 (+291.46%)
Mutual labels:  cnn, gan
Deeplearning
深度学习入门教程, 优秀文章, Deep Learning Tutorial
Stars: ✭ 6,783 (+8171.95%)
Mutual labels:  cnn, gan
Deblurgan Tf
Unofficial tensorflow (tf) implementation of DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks
Stars: ✭ 60 (-26.83%)
Mutual labels:  cnn, gan
Hand Detection.pytorch
FaceBoxes for hand detection in PyTorch
Stars: ✭ 76 (-7.32%)
Mutual labels:  cnn
Dispnet Flownet Docker
Dockerfile and runscripts for DispNet and FlowNet1 (estimation of disparity and optical flow)
Stars: ✭ 78 (-4.88%)
Mutual labels:  cnn
Tools To Design Or Visualize Architecture Of Neural Network
Tools to Design or Visualize Architecture of Neural Network
Stars: ✭ 1,143 (+1293.9%)
Mutual labels:  cnn

SEGAN

A PyTorch implementation of SEGAN based on INTERSPEECH 2017 paper SEGAN: Speech Enhancement Generative Adversarial Network.

Requirements

conda install pytorch torchvision -c pytorch
  • librosa
pip install librosa

Datasets

The clear and noisy speech datasets are downloaded from DataShare. Download the 56kHZ train datasets and test datasets, then extract them into data directory.

If you want using other datasets, you should change the path of data defined on data_preprocess.py.

Usage

Data Pre-process

python data_preprocess.py

The pre-processed datas are on data/serialized_train_data and data/serialized_test_data.

Train Model and Test

python main.py ----batch_size 128 --num_epochs 300
optional arguments:
--batch_size             train batch size [default value is 50]
--num_epochs             train epochs number [default value is 86]

The test results are on results.

Test Audio

python test_audio.py ----file_name p232_160.wav --epoch_name generator-80.pkl
optional arguments:
--file_name              audio file name
--epoch_name             generator epoch name

The generated enhanced audio is on the same directory of input audio.

Results

The example results and the pre-train Generator weight can be downloaded from BaiduYun(access code:tzdd).

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].