source: src/spectral/specdesc.h @ b9c3cd2

feature/autosinkfeature/constantqfeature/pitchshiftfeature/pydocstringsfeature/timestretchpitchshiftsamplertimestretchyinfft+
Last change on this file since b9c3cd2 was b9c3cd2, checked in by Paul Brossier <piem@piem.org>, 11 years ago

src/spectral/specdesc.h: add documentation about spectral shape descriptors

  • Property mode set to 100644
File size: 6.1 KB
Line 
1/*
2  Copyright (C) 2003-2009 Paul Brossier <piem@aubio.org>
3
4  This file is part of aubio.
5
6  aubio is free software: you can redistribute it and/or modify
7  it under the terms of the GNU General Public License as published by
8  the Free Software Foundation, either version 3 of the License, or
9  (at your option) any later version.
10
11  aubio is distributed in the hope that it will be useful,
12  but WITHOUT ANY WARRANTY; without even the implied warranty of
13  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
14  GNU General Public License for more details.
15
16  You should have received a copy of the GNU General Public License
17  along with aubio.  If not, see <http://www.gnu.org/licenses/>.
18
19*/
20
21/** \file
22 
23  Spectral description functions
24 
25  All of the following spectral description functions take as arguments the FFT
26  of a windowed signal (as created with aubio_pvoc). They output one smpl_t per
27  buffer and per channel (stored in a vector of size [channels]x[1]).
28 
29  A list of the spectral description methods currently available follows.
30
31  \section onsetdesc Onset detection functions
32
33  These functions are designed to raise at notes attacks in music signals.
34
35  \b \p energy : Energy based onset detection function
36 
37  This function calculates the local energy of the input spectral frame.
38 
39  \b \p hfc : High Frequency Content onset detection function
40 
41  This method computes the High Frequency Content (HFC) of the input spectral
42  frame. The resulting function is efficient at detecting percussive onsets.
43
44  Paul Masri. Computer modeling of Sound for Transformation and Synthesis of
45  Musical Signal. PhD dissertation, University of Bristol, UK, 1996.
46
47  \b \p complex : Complex Domain Method onset detection function
48 
49  Christopher Duxbury, Mike E. Davies, and Mark B. Sandler. Complex domain
50  onset detection for musical signals. In Proceedings of the Digital Audio
51  Effects Conference, DAFx-03, pages 90-93, London, UK, 2003.
52
53  \b \p phase : Phase Based Method onset detection function
54
55  Juan-Pablo Bello, Mike P. Davies, and Mark B. Sandler. Phase-based note onset
56  detection for music signals. In Proceedings of the IEEE International
57  Conference on Acoustics Speech and Signal Processing, pages 441­444,
58  Hong-Kong, 2003.
59
60  \b \p specdiff : Spectral difference method onset detection function
61
62  Jonhatan Foote and Shingo Uchihashi. The beat spectrum: a new approach to
63  rhythm analysis. In IEEE International Conference on Multimedia and Expo
64  (ICME 2001), pages 881­884, Tokyo, Japan, August 2001.
65
66  \b \p kl : Kullback-Liebler onset detection function
67 
68  Stephen Hainsworth and Malcom Macleod. Onset detection in music audio
69  signals. In Proceedings of the International Computer Music Conference
70  (ICMC), Singapore, 2003.
71
72  \b \p mkl : Modified Kullback-Liebler onset detection function
73
74  Paul Brossier, ``Automatic annotation of musical audio for interactive
75  systems'', Chapter 2, Temporal segmentation, PhD thesis, Centre for Digital
76  music, Queen Mary University of London, London, UK, 2006.
77
78  \b \p specflux : Spectral Flux
79
80  Simon Dixon, Onset Detection Revisited, in ``Proceedings of the 9th
81  International Conference on Digital Audio Effects'' (DAFx-06), Montreal,
82  Canada, 2006.
83
84  \section shapedesc Spectral shape descriptors
85
86  The following descriptors are described in:
87
88  Geoffroy Peeters, <i>A large set of audio features for sound description
89  (similarity and classification) in the CUIDADO project</i>, CUIDADO I.S.T.
90  Project Report 2004 (<a
91  href="http://www.ircam.fr/anasyn/peeters/ARTICLES/Peeters_2003_cuidadoaudiofeatures.pdf">pdf</a>)
92
93  \b \p centroid : Spectral centroid
94
95  The spectral centroid represents the barycenter of the spectrum.
96
97  \e Note: This function returns the result in bin. To get the spectral
98  centroid in Hz, aubio_bintofreq() should be used.
99
100  \b \p spread : Spectral spread
101
102  The spectral spread is the variance of the spectral distribution around its
103  centroid.
104
105  See also <a href="http://en.wikipedia.org/wiki/Standard_deviation">Standard
106  deviation</a> on Wikipedia.
107
108  \b \p skewness : Spectral skewness
109
110  Similarly, the skewness is computed from the third order moment of the
111  spectrum. A negative skewness indicates more energy on the lower part of the
112  spectrum. A positive skewness indicates more energy on the high frequency of
113  the spectrum.
114
115  See also <a href="http://en.wikipedia.org/wiki/Skewness">Skewness</a> on
116  Wikipedia.
117
118  \b \p kurtosis : Spectral kurtosis
119
120  The kurtosis is a measure of the flatness of the spectrum, computed from the
121  fourth order moment.
122
123  See also <a href="http://en.wikipedia.org/wiki/Kurtosis">Kurtosis</a> on
124  Wikipedia.
125
126  \b \p decrease : Spectral slope
127
128  The spectral slope represents decreasing rate of the spectral amplitude,
129  computed using a linear regression.
130
131  \b \p decrease : Spectral decrease
132
133  The spectral decrease is another representation of the decreasing rate, 
134  based on perceptual criteria.
135
136  \b \p rolloff : Spectral roll-off
137
138  This function returns the bin number below which 95% of the spectrum energy
139  is found.
140
141*/
142
143
144#ifndef ONSETDETECTION_H
145#define ONSETDETECTION_H
146
147#ifdef __cplusplus
148extern "C" {
149#endif
150
151/** spectral description structure */
152typedef struct _aubio_specdesc_t aubio_specdesc_t;
153
154/** execute spectral description function on a spectral frame
155
156  Generic function to compute spectral detescription.
157 
158  \param o spectral description object as returned by new_aubio_specdesc()
159  \param fftgrain input signal spectrum as computed by aubio_pvoc_do
160  \param desc output vector (one sample long, to send to the peak picking)
161
162*/
163void aubio_specdesc_do (aubio_specdesc_t * o, cvec_t * fftgrain,
164    fvec_t * desc);
165
166/** creation of a spectral description object
167
168  \param method spectral description method
169  \param buf_size length of the input spectrum frame
170  \param channels number of input channels
171
172*/
173aubio_specdesc_t *new_aubio_specdesc (char_t * method, uint_t buf_size,
174    uint_t channels);
175
176/** deletion of a spectral descriptor
177
178  \param o spectral descriptor object as returned by new_aubio_specdesc()
179
180*/
181void del_aubio_specdesc (aubio_specdesc_t * o);
182
183#ifdef __cplusplus
184}
185#endif
186
187#endif /* ONSETDETECTION_H */
Note: See TracBrowser for help on using the repository browser.