Readme.md

  ____   ____    _    ____  _____                          _      _     
 |  _ \ / ___|  / \  / ___|| ____|     _ __ ___   ___   __| | ___| |___ 
 | | | | |     / _ \ \___ \|  _| _____| '_ ` _ \ / _ \ / _` |/ _ \ / __|
 | |_| | |___ / ___ \ ___) | |__|_____| | | | | | (_) | (_| |  __/ \__ \
 |____/ \____/_/   \_\____/|_____|    |_| |_| |_|\___/ \__,_|\___|_|___/
                                                                       

Notebooks

This folder contains IPython / Jupyter interactive notebooks to demonstrate DCASE-models.

• Basics: examples on how to perform basic tasks.
  • Downloading datasets
  • Feature extraction
  • Data Augmentation
  • Training and evaluating a model
• Challenges: DCASE challenge examples.
  • 2020 Task 1. Acoustic scene clasification. This notebook also shows how to define a model.
• Papers: replicating paper results.
  • SB_CNN Deep Convolutional Neural Networks and Data Augmentation For Environmental Sound Classification J. Salamon and J. P. Bello IEEE Signal Processing Letters, 24(3), pages 279 - 283, 2017. This notebook includes data augmentation.
  • SB_CNN_SED Scaper: A Library for Soundscape Synthesis and Augmentation J. Salamon, D. MacConnell, M. Cartwright, P. Li, and J. P. Bello. In IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), New Paltz, NY, USA, Oct. 2017.
  • A_CRNN Sound event detection using spatial features and convolutional recurrent neural network S. Adavanne, P. Pertilä, T. Virtanen ICASSP 2017

Usage

Datasets should be stored on /datasets.

Notebooks are designed to be self-contained, but datasets can be downloaded beforehand as shown on /basics/download_and_prepare_datasets.ipynb.

Basics notebooks can be run sequentially as a tutorial.

Default parameters for each model/dataset are stored in parameters.json on the root directory. Notebooks provide examples on how to modify these parameters.

Each notebook is stored on a folder containing weights of the trained model, which can be loaded if needed.