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.

This Fork

Focused on the Python function only
Update code to work with Tensorflow 2.x
Switch off default text and tab export
Instead of an array of file paths, pass one file path or signal data
Expose sample rate and number of channels (original library assumed 44100Hz and Mono)

Installation

pip install git+https://github.com/brunoimbrizi/ADTLib.git

Usage

Python function

ADT(data, sample_rate=None, num_channels=1, text='no', tab='no', save_dir=None, output_act='no')

Name	Description	Default
`data`	Signal data or file name or file handle.
`sample_rate`	Desired sample rate of the signal [Hz], or 'None' to return the signal in its original rate.	`22050`
`num_channels`	Reduce or expand the signal to `num_channels` channels, or 'None' to return the signal with its original channels.	`1`

Other parameters: refer to the original repository.

Example

from ADTLib import ADT

out = ADT('drums.wav')

from ADTLib import ADT
import librosa

y, sr = librosa.load(path='drums.mp3')
out = ADT(y, sample_rate=sr)

License / References / Help

Refer to the original repository.