musicnn 0.0.1

Pronounced as "musician", musicnn is a set of pre-trained deep convolutional neural networks for music audio tagging

musicnn

Pronounced as "musician", musicnn is a set of pre-trained deep convolutional neural networks for music audio tagging.

Check the documentation and some basic / advanced examples for additional ideas on how to use musicnn.

Do you have questions? Check the FAQs.

Installation

pip install musicnn

or from source (to get all the documentation, including the Jupyter Notebooks):

git clone https://github.com/jordipons/musicnn.git

python setupy.py install

Predict tags

From within python, you can estimate the topN tags:

from musicnn.tagger import top_tags
top_tags('./audio/joram-moments_of_clarity-08-solipsism-59-88.mp3', model='MTT', topN=10)

['techno', 'electronic', 'synth', 'fast', 'beat', 'drums', 'no vocals', 'no vocal', 'dance', 'ambient']

Let's try another song!

top_tags('./audio/TRWJAZW128F42760DD_test.mp3', model='MTT', topN=3)

['guitar', 'piano', 'fast']

From the command-line, print to the topN tags on the screen:

python -m musicnn.tagger file_name.wav --topN 10 --print_tags
python -m musicnn.tagger file_name.au --topN 3 --print_tags

or save to a file:

python -m musicnn.tagger file_name.wav --topN 10 --output out.tags
python -m musicnn.tagger file_name.mp3 --topN 3 --output out.tags

Extract the Taggram

You can also compute the taggram using python (see our basic example for more details on how to depict it):

from musicnn.extractor import extractor
taggram, tags = extractor('./audio/joram-moments_of_clarity-08-solipsism-59-88.mp3', model='MTT')

The analyzed music clips are included in the ./audio/ folder of this repository.

You can listen to those and evaluate musicnn yourself!