2 aubioonset - a command line tool to extract musical onset times
7 aubioonset [[-i] source] [-o sink]
8 [-r rate] [-B win] [-H hop]
16 aubioonset attempts to detect onset times, the beginning of discrete sound
17 events, in audio signals.
19 When started with an input source (-i/--input), the detected onset times are
20 given on the console, in seconds.
22 When started without an input source, or with the jack option (-j/--jack),
23 aubioonset starts in jack mode.
27 This program follows the usual GNU command line syntax, with long options
28 starting with two dashes (--). A summary of options is included below.
30 -i, --input source Run analysis on this audio file. Most uncompressed and
31 compressed are supported, depending on how aubio was built.
33 -o, --output sink Save results in this file. The file will be created on
34 the model of the input file. Onset times are marked by a short wood-block
37 -r, --samplerate rate Fetch the input source, resampled at the given
38 sampling rate. The rate should be specified in Hertz as an integer. If 0,
39 the sampling rate of the original source will be used. Defaults to 0.
41 -B, --bufsize win The size of the buffer to analyze, that is the length
42 of the window used for spectral and temporal computations. Defaults to 512.
44 -H, --hopsize hop The number of samples between two consecutive analysis.
47 -O, --onset method The onset detection method to use. See ONSET METHODS
48 below. Defaults to 'default'.
50 -t, --onset-threshold thres Set the threshold value for the onset peak
51 picking. Values are typically in the range [0.001, 0.900]. Lower threshold
52 values imply more onsets detected. Increasing this threshold should reduce
53 the number of incorrect detections. Defaults to 0.3.
55 -M, --minioi value Set the minimum inter-onset interval, in seconds, the
56 shortest interval between two consecutive onsets. Defaults to 0.020
58 -s, --silence sil Set the silence threshold, in dB, under which the onset
59 will not be detected. A value of -20.0 would eliminate most onsets but the
60 loudest ones. A value of -90.0 would select all onsets. Defaults to -90.0.
62 -T, --timeformat format Set time format (samples, ms, seconds). Defaults to
65 -m, --mix-input Mix source signal to the output signal before writing to
68 -f, --force-overwrite Overwrite output file if it already exists.
70 -j, --jack Use Jack input/output. You will need a Jack connection
71 controller to feed aubio some signal and listen to its output.
73 -h, --help Print a short help message and exit.
75 -v, --verbose Be verbose.
79 Available methods are:
81 default Default distance, currently hfc
83 Default: 'default' (currently set to hfc)
85 energy Energy based distance
87 This function calculates the local energy of the input spectral frame.
89 hfc High-Frequency content
91 This method computes the High Frequency Content (HFC) of the input
92 spectral frame. The resulting function is efficient at detecting
95 Paul Masri. Computer modeling of Sound for Transformation and Synthesis of
96 Musical Signal. PhD dissertation, University of Bristol, UK, 1996.
98 complex Complex domain onset detection function
100 This function uses information both in frequency and in phase to determine
101 changes in the spectral content that might correspond to musical onsets.
102 It is best suited for complex signals such as polyphonic recordings.
104 Christopher Duxbury, Mike E. Davies, and Mark B. Sandler. Complex domain
105 onset detection for musical signals. In Proceedings of the Digital Audio
106 Effects Conference, DAFx-03, pages 90-93, London, UK, 2003.
108 phase Phase based onset detection function
110 This function uses information both in frequency and in phase to determine
111 changes in the spectral content that might correspond to musical onsets. It
112 is best suited for complex signals such as polyphonic recordings.
114 Juan-Pablo Bello, Mike P. Davies, and Mark B. Sandler. Phase-based note
115 onset detection for music signals. In Proceedings of the IEEE International
116 Conference on Acoustics Speech and Signal Processing, pages 441444,
119 specdiff Spectral difference onset detection function
121 Jonhatan Foote and Shingo Uchihashi. The beat spectrum: a new approach to
122 rhythm analysis. In IEEE International Conference on Multimedia and Expo
123 (ICME 2001), pages 881884, Tokyo, Japan, August 2001.
125 kl Kulback-Liebler onset detection function
127 Stephen Hainsworth and Malcom Macleod. Onset detection in music audio
128 signals. In Proceedings of the International Computer Music Conference
129 (ICMC), Singapore, 2003.
131 mkl Modified Kulback-Liebler onset detection function
133 Paul Brossier, ``Automatic annotation of musical audio for interactive
134 systems'', Chapter 2, Temporal segmentation, PhD thesis, Centre for
135 Digital music, Queen Mary University of London, London, UK, 2006.
137 specflux Spectral flux
139 Simon Dixon, Onset Detection Revisited, in ``Proceedings of the 9th
140 International Conference on Digital Audio Effects'' (DAFx-06), Montreal,
155 This manual page was written by Paul Brossier <piem@aubio.org>. Permission is
156 granted to copy, distribute and/or modify this document under the terms of
157 the GNU General Public License as published by the Free Software Foundation,
158 either version 3 of the License, or (at your option) any later version.