Couple learning on baseline of DCASE 2020 task 4
  • Python 94.3%
  • Shell 5.7%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2026-10-21 20:40:08 -04:00
baseline Adding num_workers to DataLoader to overcome the pb of high usage CPU 2020-06-04 16:24:24 +02:00
data_generation Adapt code to new desed update. 2021-02-27 01:53:47 +01:00
dataset Initial commit, first data 2020-03-07 04:22:24 +01:00
figures refactor measures and delete unused eval2018 + add roc curves 2020-04-09 22:49:26 +02:00
PLG-MT_run feat(crnn): add zero-float p/q Q16 GLU context gating, attention-pooled SED & BFP-64 Mel frames (+ fix missing raise) 2026-10-21 20:40:08 -04:00
scripts Regenerate the sythetic data from jams only (to avoid bugs) 2020-04-22 14:39:14 +02:00
sound-separation Merge commit '6a10a9b306' 2020-12-11 10:11:49 +01:00
tests feat(crnn): add zero-float p/q Q16 GLU context gating, attention-pooled SED & BFP-64 Mel frames (+ fix missing raise) 2026-10-21 20:40:08 -04:00
.gitignore Update .gitignore 2022-02-23 15:44:22 +08:00
conda_create_environment.sh 🔉 Release final baseline code with Sound separation evaluation 2020-04-09 00:53:20 +02:00
README.md Update README.md 2022-03-09 15:07:21 +08:00
validate_submissions.py Add evaluation set link and validate_submissions.py script 2020-05-31 12:44:28 +02:00

Couple Learning for SED

Couple Learning model

More info in the PLG-MT_run folder.

Reproducing the results of paper

See PLG-MT_run folder.

Dependencies

Python >= 3.6, pytorch >= 1.0, cudatoolkit>=9.0, pandas >= 0.24.1, scipy >= 1.2.1, pysoundfile >= 0.10.2, scaper >= 1.3.5, librosa >= 0.6.3, youtube-dl >= 2019.4.30, tqdm >= 4.31.1, ffmpeg >= 4.1, dcase_util >= 0.2.5, sed-eval >= 0.2.1, psds-eval >= 0.1.0, desed >= 1.3.0

A simplified installation procedure example is provided below for python 3.6 based Anconda distribution for Linux based system:

  1. install Ananconda
  2. launch conda_create_environment.sh (recommended line by line)

Dataset

All the scripts to get the data (soundbank, generated, separated) are in the scripts folder and they use python files from data_generation folder.

Scripts to generate the dataset

In the scripts/ folder, you can find the different steps to:

  • Download recorded data and synthetic material.
  • Generate synthetic soundscapes
  • Reverberate synthetic data (Not used in the baseline)
  • Separate sources of recorded and synthetic mixtures

It is likely that you'll have download issues with the real recordings. At the end of the download, please send a mail with the TSV files created in the missing_files directory.

However, if none of the audio files have been downloaded, it is probably due to an internet, proxy problem. See Desed repo or Desed_website for more info.

Base dataset

The dataset for sound event detection of DCASE2020 task 4 is composed of:

  • Train:
    • *weak (DESED, recorded, 1 578 files)
    • *unlabel_in_domain (DESED, recorded, 14 412 files)
    • synthetic soundbank (DESED, synthetic, 2060 background (SINS only) + 1006 foreground files)
  • *Validation (DESED, recorded, 1 168 files):
    • test2018 (288 files)
    • eval2018 (880 files)

Baselines dataset

SED baseline
  • Train:
    • weak
    • unlabel_in_domain
    • synthetic20/soundscapes (separated in train/valid-80%/20%)
  • Validation:
    • validation