DCASE2024 Task 9 Baseline System

The code of this repository is mostly from AudioSep. If you use this code, please cite the original repository following:

@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}
}

@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},
}

Setup

Clone the repository and setup the conda environment:

  git clone https://github.com/Audio-AGI/dcase2024_task9_baseline.git && \
  cd dcase2024_task9_baseline && \ 
  conda env create -f environment.yml && \
  conda activate AudioSep

Download CLAP (music_speech_audioset_epoch_15_esc_89.98.pt ) model weight at checkpoint/, which is necessary for training AudioSep model.

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:

data:
    datafiles:
        - 'datafiles/your_datafile_1.json'
        - 'datafiles/your_datafile_2.json'
        ...

Train AudioSep from scratch:

  python train.py --workspace workspace/AudioSep --config_yaml config/audiosep_base.yaml --resume_checkpoint_path ''

Finetune AudioSep from pretrained checkpoint:

  python train.py --workspace workspace/AudioSep --config_yaml config/audiosep_base.yaml --resume_checkpoint_path path_to_checkpoint

The development set including Clotho v2 and augmented FSD50k datasets can be found here.

Baseline model

For the baseline model, we trained the AudioSep model using Clotho and augmented FSD50K datasets for 200k steps with a batch size of 16 using one Nvidia A100 GPU (around 1 day). The checkpoint of the baseline model can be found here.

Evaluation

• Download the audio files and the indexes of the synthetic validation data from Zenodo.

# paired data indexes for evaluation
- lass_synthetic_validation.csv
# validation audio files
- lass_validation:
    - audio_file_1.wav
    ...

• Modify the paths as shown below and then run the evaluation script dcase_evaluator.py.

dcase_evaluator = DCASEEvaluator(
        sampling_rate=16000,
        eval_indexes='lass_synthetic_validation.csv',
        audio_dir='lass_validation'
    )

checkpoint_path = 'audiosep,baseline-16k,step=200000.ckpt'
eval(dcase_evaluator, checkpoint_path)

"""
-------  Start Evaluation  -------
Evaluation on DCASE T9 synthetic validation set.
Results: SDR: SDR: 5.708, SDRi: 5.673, SISDR: 3.862
------------- Done ---------------
"""