Skip to content

QutEcoacoustics/recognizer_workshop

Repository files navigation

Getting Started

Requirements

Python with jupyter notebooks installed. Other packages are installed as part of the notebook.

Launch the notebook

In terminal, run jupyter notebook

Included models

artifacts/models contains models for detecting some species. Currently these are

  • Kookaburra Attribution: QUT Ecoacoustics Group
  • Noisy Pitta (98% accuracy on test set) Attribution: QUT Ecoacoustics Group
  • Rufous Scrub Bird Attribution: Brian Hawkins, QUT Ecoacoustics Group
  • Accuracy is on the test set we have access to. This does not necessarily represent the accuracy audio from different locations and recordings

License

Models and code are licensed under a Creative Commons Attribution 4.0 International License. https://choosealicense.com/licenses/cc-by-sa-4.0/