You can install with pip: $ pip install hybrid-vocal-classifier
For more detail, please see: https://hybrid-vocal-classifier.readthedocs.io/en/latest/install.html#install
To learn how to use hybrid-vocal-classifier, please see the documentation at:
http://hybrid-vocal-classifier.readthedocs.io
You can find a tutorial here: https://hybrid-vocal-classifier.readthedocs.io/en/latest/tutorial.html
A more interactive tutorial in Jupyter notebooks is here:
https://github.com/NickleDave/hybrid-vocal-classifier-tutorial
the hybrid-vocal-classifier library (hvc for short)
makes it easier for researchers studying
animal vocalizations and bioacoustics
to apply machine learning algorithms to their data.
The focus on automating the sort of annotations
often used by researchers studying
vocal learning
sets hvc apart from more general software tools for bioacoustics.
In addition to automating annotation of data,
hvc aims to make it easy for you to compare different models people have proposed,
using the data you have in your lab,
so you can see for yourself which one works best for your needs.
A related goal is to help you figure out
just how much data you have to label to get "good enough" accuracy for your analyses.
You can think of hvc as a high-level wrapper around
the scikit-learn library,
plus built-in functionality for working with annotated animal sounds.
If you are having issues, please let us know.
- Issue Tracker: https://github.com/NickleDave/hybrid-vocal-classifier/issues
- Issue Tracker: https://github.com/NickleDave/hybrid-vocal-classifier/issues
- Source Code: https://github.com/NickleDave/hybrid-vocal-classifier
You can see project history and work in progress in the CHANGELOG
The project is licensed under the BSD license.
If you use this library, please cite its DOI:
hvc was originally developed in the Sober lab
as a tool to automate annotation of birdsong (as shown in the picture above).
It grew out of a submission to the
SciPy 2016 conference
and later developed into a library,
as presented in this talk: https://youtu.be/BwNeVNou9-s