nussl External File Zoo
This website houses and distributes extra files that are too large to
nussl, such as trained deep learning models, audio files,
and benchmark files for testing. Please see the contributions section
on Github to contribute files associated with your algorithms.
The best and easiest way to download these files is through nussl
itself. The download functions in
nussl will download the desired
file from this website (if the file is not already on your computer)
and when finished the functions will return the full path to the
You can see what files are available through the helper
nussl.utils. For instance, to see what
audio files are available simply do:
>>> import nussl
File Name Duration (sec) Size Description
dev1_female3_inst_mix.wav 10.0 1.7MiB Instantaneous mixture of three female speakers talking in a stereo field.
dev1_female3_synthconv_130ms_5cm_mix.wav 10.0 1.7MiB Three female speakers talking in a stereo field, with 130ms of inter-channel delay.
K0140.wav 5.0 431.0KiB Acoustic piano playing middle C.
K0149.wav 5.0 430.0KiB Acoustic piano playing the A above middle C. (A440)
This message has the same information that is shown on this website.
To download one of these files, just give a file name to the associated download function like so:
>>> audio_path = nussl.utils.download_audio_file('K0140.wav')
Saving file at ~/.nussl/audio/K0140.wav
Downloading K0140.wav from https://nussl.ci.northwestern.edu/static/audio/K0140.wav
The download function returns a path to the audio file.
There are analogous for downloading trained models and benchmarks. These functions work the same as the audio example above; you can see what files are available and descriptions of them, as well as download them. For example, to download a model do the following:
>>> model_path = nussl.utils.download_trained_model('example.model')
Saving file at ~/.nussl/models/example.model
Downloading example.model from https://nussl.ci.northwestern.edu/static/trained_models/example.model
|Acoustic piano playing middle C.
|Acoustic piano playing A above middle C (A440).
|Instantaneous mixture of three female speakers talking in a stereo field.
|Three female speakers talking in a stereo field, with 130ms of inter-channel delay.
|Drums and flute.
|Drums playing a simple beat.
|Flute playing a melody repeated twice (with slightly altered articulation).
|Mixture of flute, drums, and female speaker.
|Mixture of flute, drums, and female speaker in stereo.
|Mixture of flute and drums.
|Mixture of flute and drums. Alt.
|Instantaneous mixture of flute and drums in a stereo field.
|Two speaker mixture from WSJ0-2mix.
|Excerpt from MUSDB - Al James - Schoolboy Fascination
|Two marimbas playing similar material with different timbre. From Bregman's Demonstrations of Auditory Scene Analysis
|F1 from DAPS dataset, recorded in a conference room.
|Impulse response taken from MIT IR dataset.
No Trained Models are available at this time.
|Background matrix for Repet class benchmark test.
|Foreground matrix for Repet class benchmark test.
|Attenuation bins numpy array for DUET benchmark test.
|Attenuation/Delay estimation numpy array for DUET benchmark test.
|Attenuation peak numpy array for DUET benchmark test.
|Delay numpy array for DUET benchmark test.
|Delay bins for DUET benchmark tests.
|Delay histogram peaks for DUET benchmarks
|Complete attenuation/delay histogram for the DUET benchmark test.
|Result masks for the DUET benchmark test.
|Selected peaks on the AD histogram for DUET benchmark test.
|STFT channel 0 for the DUET benchmark test.
|STFT channel 1 for the DUET benchmark test.
|Symmetric attenuation histogram for the DUET benchmark test.
|Frequency matrix for the DUET benchmark test.
|A small dataset for testing model training pipelines and separation quality.
|Small dataset for testing music separation pipeline.