https://github.com/Audio-AGI/AudioSep
Official implementation of "Separate Anything You Describe"
https://github.com/Audio-AGI/AudioSep
Last synced: 4 months ago
JSON representation
Official implementation of "Separate Anything You Describe"
- Host: GitHub
- URL: https://github.com/Audio-AGI/AudioSep
- Owner: Audio-AGI
- License: mit
- Created: 2023-08-08T19:18:47.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-03-31T14:00:53.000Z (11 months ago)
- Last Synced: 2024-10-22T16:41:08.915Z (4 months ago)
- Language: Python
- Homepage: https://audio-agi.github.io/Separate-Anything-You-Describe/
- Size: 16.1 MB
- Stars: 1,604
- Watchers: 64
- Forks: 117
- Open Issues: 24
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# Separate Anything You Describe
[](https://arxiv.org/abs/2308.05037) [](https://github.com/Audio-AGI/AudioSep/) [](https://audio-agi.github.io/Separate-Anything-You-Describe) [](https://colab.research.google.com/github/Audio-AGI/AudioSep/blob/main/AudioSep_Colab.ipynb) [](https://huggingface.co/spaces/Audio-AGI/AudioSep) [](https://replicate.com/cjwbw/audiosep)
This repository contains the official implementation of ["Separate Anything You Describe"](https://audio-agi.github.io/Separate-Anything-You-Describe/AudioSep_arXiv.pdf).
We introduce AudioSep, a foundation model for open-domain sound separation with natural language queries. AudioSep demonstrates strong separation performance and impressive zero-shot generalization ability on numerous tasks, such as audio event separation, musical instrument separation, and speech enhancement. Check out the separated audio examples on the [Demo Page](https://audio-agi.github.io/Separate-Anything-You-Describe/)!
## Setup
Clone the repository and setup the conda environment:
```shell
git clone https://github.com/Audio-AGI/AudioSep.git && \
cd AudioSep && \
conda env create -f environment.yml && \
conda activate AudioSep
```
Download [model weights](https://huggingface.co/spaces/Audio-AGI/AudioSep/tree/main/checkpoint) at `checkpoint/`.
If you're using this checkpoint for the DCASE 2024 Task 9 challenge participation, please note that this checkpoint was trained using audio at 32k Hz, with a window size of 2048 points and a hop size of 320 points in the STFT operation, which is different with the challenge baseline system provided (16k Hz, window size 1024, hop size 160).
## Inference
```python
from pipeline import build_audiosep, inference
import torch
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
model = build_audiosep(
config_yaml='config/audiosep_base.yaml',
checkpoint_path='checkpoint/audiosep_base_4M_steps.ckpt',
device=device)
audio_file = 'path_to_audio_file'
text = 'textual_description'
output_file='separated_audio.wav'
# AudioSep processes the audio at 32 kHz sampling rate
inference(model, audio_file, text, output_file, device)
```
To load directly from Hugging Face, you can do the following:
```python
from models.audiosep import AudioSep
from utils import get_ss_model
import torch
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
ss_model = get_ss_model('config/audiosep_base.yaml')
model = AudioSep.from_pretrained("nielsr/audiosep-demo", ss_model=ss_model)
audio_file = 'path_to_audio_file'
text = 'textual_description'
output_file='separated_audio.wav'
# AudioSep processes the audio at 32 kHz sampling rate
inference(model, audio_file, text, output_file, device)
```
Use chunk-based inference to save memory:
```python
inference(model, audio_file, text, output_file, device, use_chunk=True)
```
## Training
To utilize your audio-text paired dataset:
1. Format your dataset to match our JSON structure. Refer to the provided template at `datafiles/template.json`.
2. Update the `config/audiosep_base.yaml` file by listing your formatted JSON data files under `datafiles`. For example:
```yaml
data:
datafiles:
- 'datafiles/your_datafile_1.json'
- 'datafiles/your_datafile_2.json'
...
```
Train AudioSep from scratch:
```python
python train.py --workspace workspace/AudioSep --config_yaml config/audiosep_base.yaml --resume_checkpoint_path checkpoint/ ''
```
Finetune AudioSep from pretrained checkpoint:
```python
python train.py --workspace workspace/AudioSep --config_yaml config/audiosep_base.yaml --resume_checkpoint_path path_to_checkpoint
```
## Benchmark Evaluation
Download the [evaluation data](https://drive.google.com/drive/folders/1PbCsuvdrzwAZZ_fwIzF0PeVGZkTk0-kL?usp=sharing) under the `evaluation/data` folder. The data should be organized as follows:
```yaml
evaluation:
data:
- audioset/
- audiocaps/
- vggsound/
- music/
- clotho/
- esc50/
```
Run benchmark inference script, the results will be saved at `eval_logs/`
```python
python benchmark.py --checkpoint_path audiosep_base_4M_steps.ckpt
"""
Evaluation Results:
VGGSound Avg SDRi: 9.144, SISDR: 9.043
MUSIC Avg SDRi: 10.508, SISDR: 9.425
ESC-50 Avg SDRi: 10.040, SISDR: 8.810
AudioSet Avg SDRi: 7.739, SISDR: 6.903
AudioCaps Avg SDRi: 8.220, SISDR: 7.189
Clotho Avg SDRi: 6.850, SISDR: 5.242
"""
```
## Cite this work
If you found this tool useful, please consider citing
```bibtex
@article{liu2023separate,
title={Separate Anything You Describe},
author={Liu, Xubo and Kong, Qiuqiang and Zhao, Yan and Liu, Haohe and Yuan, Yi, and Liu, Yuzhuo, and Xia, Rui and Wang, Yuxuan, and Plumbley, Mark D and Wang, Wenwu},
journal={arXiv preprint arXiv:2308.05037},
year={2023}
}
```
```bibtex
@inproceedings{liu22w_interspeech,
title={Separate What You Describe: Language-Queried Audio Source Separation},
author={Liu, Xubo and Liu, Haohe and Kong, Qiuqiang and Mei, Xinhao and Zhao, Jinzheng and Huang, Qiushi, and Plumbley, Mark D and Wang, Wenwu},
year=2022,
booktitle={Proc. Interspeech},
pages={1801--1805},
}
```
## Contributors :
