Deep Denoising Autoencoder (DDAE) for Speech Enhancement
Tensorflow implementation of Speech Enhancement Based on Deep Denoising Autoencoder
Getting Started
Clone This repository to your local machine and run create_dir.sh first.
Prerequisites
- python 3.5
- tensorflow-gpu 1.8.0
- scikit-learn 0.19.1
- scipy 1.1.0
- h5py 2.7.1
- librosa 0.5.1
- numpy 1.14.3
- tqdm 4.23.2
Getting Started
- Download free dataset from VoxForge for clean data. Here I would recommed download cmu_us_awb_arctic.tgz
- Unzip clean dataset to /DeepDenoisingAutoencoder/data/raw/clean/
- Download free dataset from ESC-50 for noise data.
- Move ESC-50-master/audio to /DeepDenoisingAutoencoder/data/raw/noise/
- Set parameters in python/main.py
- Run python/main.py
Result
Deployment
You can read many comments inside all .py files.
Authors
Yu-Ding Lu - Linkedin
License
This project is licensed under the MIT License - see the LICENSE.md file for details
Acknowledgments
- Bio-ASP lab - CITI - Academia Sinica