Skip to main content

C++ / Python library with signal processing and tempo estimation utilities.

Project description


C++ / Python library for signal processing and tempo information extraction.


MP3, FLAC and WAV audio formats are supported.




libtempo [audio] {OPTIONS}

      -h, --help                        Display the help menu
        --preferred_bpm=[preferred_bpm]   (Default: 130) BPM around which the
                                          real bpm will be chosen.
                                          (Default: 0.500000) Precision of the
                                          BPM before correction..
        --smooth_length=[smooth_length]   (Default: 20.000000) Length in seconds
                                          over which the tempogram will be
                                          stabilized to extract a steady tempo.
                                          (Default: 10.000000) Minimum length
                                          for a tempo section in seconds.
                                          (Default: 40.000000) Maximum section
                                          length in seconds after which section
                                          is split in half.
        --bpm_window=[bpm_window]         BPM range which to check for tempo
                                          Tempo multiples to consider when
                                          searching for correct offset
        --ref_tempo=[ref_tempo]           (Default: 60) Reference tempo defining
                                          the partition of BPM into tempo
                                          octaves for calculating cyclic
        --tempo_window=[tempo_window]     (Default: 8) Analysis window length in
                                          seconds for calculating tempogram.
        --octave_divider=[octave_divider] (Default: 120) Number of tempo classes
                                          used for representing a tempo octave.
                                          This parameter controls the
                                          dimensionality of cyclic tempogram.
        --triplet_weight=[triplet_weight] (Default: 3.000000) Weight of the
                                          triplet intensity which will be adeed
                                          to its base intensity.
                                          (Default: 2.000000) Window around
                                          candidate bpm which to search for a
                                          more fine and correct bpm.
        --bpm_doubt_step=[bpm_doubt_step] (Default: 0.100000) Steps which to
                                          take inside the doubt window to fine
                                          tune the bpm.
                                          (Default: 8) Click subdivision for the
                                          click track.
      Output Flags
                                          (Default: 1) Wether or not a click
                                          track should be generated
        -o, --osu                         Wether or not to generate tempo data
                                          in osu format.
        --viz                             Saves a html file with useful graphs.
        -d, --dump                        Dump generated tempograms and novelty
                                          curves into a file.
      audio                             Audio file to extract tempo of.
      "--" can be used to terminate flag options and force all following
      arguments to be treated as positional options




Flaws / TODO's

  • BPM calculated is the percieved bpm. There is also real bpm which may be twice as high. (I would love to hear any suggestions how this could be helped)
  • Pars where bpm is constantly changed are rounded away when making secions.
  • Sometimes offset might be shifted by a 1/4th or 1/8th note.


Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

libtempo_py-1.0.0-cp36-cp36m-win_amd64.whl (213.9 kB view hashes)

Uploaded CPython 3.6m Windows x86-64

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page