CNNOnsetDetector

#!/usr/bin/env python
# encoding: utf-8
"""
CNNOnsetDetector onset detection algorithm.

"""

from __future__ import absolute_import, division, print_function

import argparse

from madmom.audio import SignalProcessor
from madmom.features import (ActivationsProcessor, CNNOnsetProcessor,
                             OnsetPeakPickingProcessor)
from madmom.io import write_onsets
from madmom.processors import IOProcessor, io_arguments


def main():
    """CNNOnsetDetector"""

    # define parser
    p = argparse.ArgumentParser(
        formatter_class=argparse.RawDescriptionHelpFormatter, description='''
    The CNNOnsetDetector program detects all onsets in an audio file with a
    convolutional neural network as described in:

    "Musical Onset Detection with Convolutional Neural Networks"
    Jan Schlüter and Sebastian Böck.
    Proceedings of the 6th International Workshop on Machine Learning and
    Music, 2013.

    The implementation follows as closely as possible the original one, but
    part of the signal pre-processing differs in minor aspects, so results can
    differ slightly, too.

    This program can be run in 'single' file mode to process a single audio
    file and write the detected onsets to STDOUT or the given output file.

      $ CNNOnsetDetector single INFILE [-o OUTFILE]

    If multiple audio files should be processed, the program can also be run
    in 'batch' mode to save the detected onsets to files with the given suffix.

      $ CNNOnsetDetector batch [-o OUTPUT_DIR] [-s OUTPUT_SUFFIX] FILES

    If no output directory is given, the program writes the files with the
    detected onsets to the same location as the audio files.

    The 'pickle' mode can be used to store the used parameters to be able to
    exactly reproduce experiments.

    ''')
    # version
    p.add_argument('--version', action='version',
                   version='CNNOnsetDetector.2016')
    # input/output options
    io_arguments(p, output_suffix='.onsets.txt')
    ActivationsProcessor.add_arguments(p)
    # signal processing arguments
    SignalProcessor.add_arguments(p, norm=False, gain=0)
    # peak picking arguments
    OnsetPeakPickingProcessor.add_arguments(p, threshold=0.54, smooth=0.05)

    # parse arguments
    args = p.parse_args()

    # set immutable defaults
    args.fps = 100
    args.pre_max = 1. / args.fps
    args.post_max = 1. / args.fps

    # print arguments
    if args.verbose:
        print(args)

    # input processor
    if args.load:
        # load the activations from file
        in_processor = ActivationsProcessor(mode='r', **vars(args))
    else:
        # use a CNN to predict the onsets
        in_processor = CNNOnsetProcessor(**vars(args))

    # output processor
    if args.save:
        # save the RNN onset activations to file
        out_processor = ActivationsProcessor(mode='w', **vars(args))
    else:
        # detect the onsets and output them
        peak_picking = OnsetPeakPickingProcessor(**vars(args))
        out_processor = [peak_picking, write_onsets]

    # create an IOProcessor
    processor = IOProcessor(in_processor, out_processor)

    # and call the processing function
    args.func(processor, **vars(args))


if __name__ == '__main__':
    main()