AudioSR: Versatile Audio Super-resolution at Scale

Pass your audio in, AudioSR will make it high fidelity!

Work on all types of audio (e.g., music, speech, dog, raining, ...) & all sampling rates.

Original Repo

Table of Contents

• Installation
• Prepare_Datasets
• Usage
• Reference

Installation

1. Create an Anaconda environment:

conda create -n audiosr python=3.9

2. Activate the environment:

conda activate audiosr

3. Clone this repository to your local machine:

git clone https://github.com/ORI-Muchim/AudioSR-Upsampling.git

4. Navigate to the cloned directory:

cd AudioSR-Upsampling

5. Install the necessary dependencies:

pip install -r requirements.txt

Prepare_Datasets

Place the audio files as follows.

.mp3 or .wav files are okay.

AudioSR-Upsampling
├────datasets
│       ├───speaker0
│       │   ├────1.mp3
│       │   └────2.mp3
│       ├───speaker1
│       │    ├───1.wav
│       │    └───1.wav
│       ├───speaker2
│       │   ├────1.wav
│       └───└────1.wav
├────audiosr
├────.gitignore
├────main.py
├────Readme.md
└────requirements.txt

This is just an example, and it's okay to add more speakers.

When you put audio datasets in one folder, please unify all the extensions into one.

Usage

python main.py

Reference

Thank you for falsewinnet for helping me create ./audiosr/__main__.py.

If you find this repo useful, please consider citing:

@article{liu2023audiosr,
  title={{AudioSR}: Versatile Audio Super-resolution at Scale},
  author={Liu, Haohe and Chen, Ke and Tian, Qiao and Wang, Wenwu and Plumbley, Mark D},
  journal={arXiv preprint arXiv:2309.07314},
  year={2023}
}