ADTLib
Automatic Drum Transcription Library in Python
3 min read
Automatic Drum Transcription Library (ADTLib)
The automatic drum transcription (ADT) library contains open source ADT algorithms to aid other researchers in areas of music information retrieval (MIR). The algorithms return both a .txt file of kick drum, snare drum, and hi-hat onsets and an automatically generated drum tabulature.
Required Packages:
- numpy
- scipy
- madmom
- tensorflow
- fpdf (for tab creation)
- Python
# Install
pip install ADTLib
# To update the library
pip install --upgrade ADTLib
Usage
- Python function
- Command line
from ADTLib import ADT
# ADT on single file
out = ADT(['Drum.wav'])
# Multiple files. Save onsets (.txt) and drum tabulature (.pdf)
Onsets=ADT(['Drum.wav', 'Drum1.wav', 'Drum2.wav'])
# Multiple files. Save drum tabulature but not onset times
Onsets=ADT('~/Drum.wav', '~/Desktop/Drum1.wav', text='no', save_dir='~/Desktop')
# Save onset times and activation functions
[Onsets, ActivationFunctions] = ADT(['Drum.wav'], output_act='yes')
# Single file
ADT Drum.wav
# Multiple files
ADT Drum.wav Drum1.wav Drum2.wav
# Save onset times as .txt and drum tabulature as .pdf
ADT -od ~/Desktop -o no ~/Drum.wav ~/Desktop/Drum1.wav
See the usage page for more information.
Browser Version (ADTWeb)
ADTWeb: a browser based version of ADTLib is now available.
References
- Southall, C., R. Stables, J. Hockman, Automatic Drum Transcription Using Bi-directional Recurrent Neural Networks, Proceedings of the 17th International Society for Music Information Retrieval Conference (ISMIR), 2016.
- Southall, C., R. Stables, J. Hockman, Automatic Drum Transcription For Polyphonic Recordings Using Soft Attention Mechanisms and Convolutional Neural Networks, Proceedings of the 18th International Society for Music Information Retrieval Conference (ISMIR), 2017.
- Southall, C., N. Jillings, R. Stables, J. Hockman, ADTWeb: An Open Source Browser Based Automatic Drum Transcription System. Proceedings of the 18th International Society for Music Information Retrieval Conference (ISMIR), 2017.