[83963b3] | 1 | /* |
---|
[0683ee2] | 2 | Copyright (C) 2003-2015 Paul Brossier <piem@aubio.org> |
---|
[83963b3] | 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 |
---|
[a17e8c9] | 22 | * various functions useful in audio signal processing |
---|
[83963b3] | 23 | */ |
---|
| 24 | |
---|
[6f42c16] | 25 | #ifndef AUBIO_MUSICUTILS_H |
---|
| 26 | #define AUBIO_MUSICUTILS_H |
---|
[83963b3] | 27 | |
---|
| 28 | #ifdef __cplusplus |
---|
| 29 | extern "C" { |
---|
| 30 | #endif |
---|
| 31 | |
---|
[0683ee2] | 32 | /** create window |
---|
[e84ab04] | 33 | |
---|
| 34 | \param window_type type of the window to create |
---|
| 35 | \param size length of the window to create (see fvec_set_window()) |
---|
[0683ee2] | 36 | |
---|
[e84ab04] | 37 | */ |
---|
| 38 | fvec_t *new_aubio_window (char_t * window_type, uint_t size); |
---|
| 39 | |
---|
| 40 | /** set elements of a vector to window coefficients |
---|
| 41 | |
---|
| 42 | \param window exsting ::fvec_t to use |
---|
| 43 | \param window_type type of the window to create |
---|
| 44 | |
---|
| 45 | List of available window types: "rectangle", "hamming", "hanning", |
---|
| 46 | "hanningz", "blackman", "blackman_harris", "gaussian", "welch", "parzen", |
---|
| 47 | "default". |
---|
| 48 | |
---|
| 49 | "default" is equivalent to "hanningz". |
---|
| 50 | |
---|
[83963b3] | 51 | References: |
---|
[0683ee2] | 52 | |
---|
[83963b3] | 53 | - <a href="http://en.wikipedia.org/wiki/Window_function">Window |
---|
| 54 | function</a> on Wikipedia |
---|
| 55 | - Amalia de Götzen, Nicolas Bernardini, and Daniel Arfib. Traditional (?) |
---|
| 56 | implementations of a phase vocoder: the tricks of the trade. In Proceedings of |
---|
| 57 | the International Conference on Digital Audio Effects (DAFx-00), pages 37–44, |
---|
| 58 | Uni- versity of Verona, Italy, 2000. |
---|
[2e69640] | 59 | (<a href="http://www.cs.princeton.edu/courses/archive/spr09/cos325/Bernardini.pdf"> |
---|
| 60 | pdf</a>) |
---|
[83963b3] | 61 | |
---|
[33cd81f] | 62 | */ |
---|
| 63 | uint_t fvec_set_window (fvec_t * window, char_t * window_type); |
---|
| 64 | |
---|
[83963b3] | 65 | /** compute the principal argument |
---|
| 66 | |
---|
| 67 | This function maps the input phase to its corresponding value wrapped in the |
---|
| 68 | range \f$ [-\pi, \pi] \f$. |
---|
| 69 | |
---|
| 70 | \param phase unwrapped phase to map to the unit circle |
---|
[0683ee2] | 71 | |
---|
[83963b3] | 72 | \return equivalent phase wrapped to the unit circle |
---|
| 73 | |
---|
| 74 | */ |
---|
| 75 | smpl_t aubio_unwrap2pi (smpl_t phase); |
---|
| 76 | |
---|
| 77 | /** convert frequency bin to midi value */ |
---|
| 78 | smpl_t aubio_bintomidi (smpl_t bin, smpl_t samplerate, smpl_t fftsize); |
---|
| 79 | |
---|
| 80 | /** convert midi value to frequency bin */ |
---|
| 81 | smpl_t aubio_miditobin (smpl_t midi, smpl_t samplerate, smpl_t fftsize); |
---|
| 82 | |
---|
| 83 | /** convert frequency bin to frequency (Hz) */ |
---|
| 84 | smpl_t aubio_bintofreq (smpl_t bin, smpl_t samplerate, smpl_t fftsize); |
---|
| 85 | |
---|
| 86 | /** convert frequency (Hz) to frequency bin */ |
---|
| 87 | smpl_t aubio_freqtobin (smpl_t freq, smpl_t samplerate, smpl_t fftsize); |
---|
| 88 | |
---|
[ff9c62a] | 89 | /** convert frequency (Hz) to mel |
---|
| 90 | |
---|
| 91 | \param freq input frequency, in Hz |
---|
| 92 | |
---|
| 93 | \return output mel |
---|
| 94 | |
---|
| 95 | Converts a scalar from the frequency domain to the mel scale using Slaney |
---|
| 96 | Auditory Toolbox's implementation: |
---|
| 97 | |
---|
| 98 | If \f$ f < 1000 \f$, \f$ m = 3 f / 200 \f$. |
---|
| 99 | |
---|
| 100 | If \f$ f >= 1000 \f$, \f$ m = 1000 + 27 \frac{{ln}(f) - ln(1000))} |
---|
| 101 | {{ln}(6400) - ln(1000)} |
---|
| 102 | \f$ |
---|
| 103 | |
---|
| 104 | See also |
---|
| 105 | -------- |
---|
| 106 | |
---|
| 107 | aubio_meltohz(), aubio_hztomel_htk(). |
---|
| 108 | |
---|
| 109 | */ |
---|
| 110 | smpl_t aubio_hztomel (smpl_t freq); |
---|
| 111 | |
---|
| 112 | /** convert mel to frequency (Hz) |
---|
| 113 | |
---|
| 114 | \param mel input mel |
---|
| 115 | |
---|
| 116 | \return output frequency, in Hz |
---|
| 117 | |
---|
| 118 | Converts a scalar from the mel scale to the frequency domain using Slaney |
---|
| 119 | Auditory Toolbox's implementation: |
---|
| 120 | |
---|
| 121 | If \f$ f < 1000 \f$, \f$ f = 200 m/3 \f$. |
---|
| 122 | |
---|
| 123 | If \f$ f \geq 1000 \f$, \f$ f = 1000 + \left(\frac{6400}{1000}\right) |
---|
| 124 | ^{\frac{m - 1000}{27}} \f$ |
---|
| 125 | |
---|
| 126 | See also |
---|
| 127 | -------- |
---|
| 128 | |
---|
| 129 | aubio_hztomel(), aubio_meltohz_htk(). |
---|
| 130 | |
---|
| 131 | References |
---|
| 132 | ---------- |
---|
| 133 | |
---|
| 134 | Malcolm Slaney, *Auditory Toolbox Version 2, Technical Report #1998-010* |
---|
| 135 | https://engineering.purdue.edu/~malcolm/interval/1998-010/ |
---|
| 136 | |
---|
| 137 | */ |
---|
| 138 | smpl_t aubio_meltohz (smpl_t mel); |
---|
| 139 | |
---|
| 140 | /** convert frequency (Hz) to mel |
---|
| 141 | |
---|
| 142 | \param freq input frequency, in Hz |
---|
| 143 | |
---|
| 144 | \return output mel |
---|
| 145 | |
---|
| 146 | Converts a scalar from the frequency domain to the mel scale, using the |
---|
| 147 | equation defined by O'Shaughnessy, as implemented in the HTK speech |
---|
| 148 | recognition toolkit: |
---|
| 149 | |
---|
| 150 | \f$ m = 1127 + ln(1 + \frac{f}{700}) \f$ |
---|
| 151 | |
---|
| 152 | See also |
---|
| 153 | -------- |
---|
| 154 | |
---|
| 155 | aubio_meltohz_htk(), aubio_hztomel(). |
---|
| 156 | |
---|
| 157 | References |
---|
| 158 | ---------- |
---|
| 159 | |
---|
| 160 | Douglas O'Shaughnessy (1987). *Speech communication: human and machine*. |
---|
| 161 | Addison-Wesley. p. 150. ISBN 978-0-201-16520-3. |
---|
| 162 | |
---|
| 163 | HTK Speech Recognition Toolkit: http://htk.eng.cam.ac.uk/ |
---|
| 164 | |
---|
| 165 | */ |
---|
| 166 | smpl_t aubio_hztomel_htk (smpl_t freq); |
---|
| 167 | |
---|
| 168 | /** convert mel to frequency (Hz) |
---|
| 169 | |
---|
| 170 | \param mel input mel |
---|
| 171 | |
---|
| 172 | \return output frequency, in Hz |
---|
| 173 | |
---|
| 174 | Converts a scalar from the mel scale to the frequency domain, using the |
---|
| 175 | equation defined by O'Shaughnessy, as implemented in the HTK speech |
---|
| 176 | recognition toolkit: |
---|
| 177 | |
---|
| 178 | \f$ f = 700 * {e}^\left(\frac{f}{1127} - 1\right) \f$ |
---|
| 179 | |
---|
| 180 | See also |
---|
| 181 | -------- |
---|
| 182 | |
---|
| 183 | aubio_hztomel_htk(), aubio_meltohz(). |
---|
| 184 | |
---|
| 185 | */ |
---|
| 186 | smpl_t aubio_meltohz_htk (smpl_t mel); |
---|
| 187 | |
---|
[83963b3] | 188 | /** convert frequency (Hz) to midi value (0-128) */ |
---|
| 189 | smpl_t aubio_freqtomidi (smpl_t freq); |
---|
| 190 | |
---|
| 191 | /** convert midi value (0-128) to frequency (Hz) */ |
---|
| 192 | smpl_t aubio_miditofreq (smpl_t midi); |
---|
| 193 | |
---|
| 194 | /** clean up cached memory at the end of program |
---|
[0683ee2] | 195 | |
---|
[83963b3] | 196 | This function should be used at the end of programs to purge all cached |
---|
| 197 | memory. So far it is only useful to clean FFTW's cache. |
---|
| 198 | |
---|
| 199 | */ |
---|
| 200 | void aubio_cleanup (void); |
---|
| 201 | |
---|
| 202 | /** zero-crossing rate (ZCR) |
---|
| 203 | |
---|
| 204 | The zero-crossing rate is the number of times a signal changes sign, |
---|
| 205 | divided by the length of this signal. |
---|
| 206 | |
---|
| 207 | \param v vector to compute ZCR from |
---|
| 208 | |
---|
| 209 | \return zero-crossing rate of v |
---|
| 210 | |
---|
| 211 | */ |
---|
| 212 | smpl_t aubio_zero_crossing_rate (fvec_t * v); |
---|
| 213 | |
---|
[4b891e9] | 214 | /** compute sound level on a linear scale |
---|
[5b41ef9] | 215 | |
---|
| 216 | This gives the average of the square amplitudes. |
---|
| 217 | |
---|
[4b891e9] | 218 | \param v vector to compute level from |
---|
[5b41ef9] | 219 | |
---|
| 220 | \return level of v |
---|
| 221 | |
---|
| 222 | */ |
---|
[e9bd022] | 223 | smpl_t aubio_level_lin (const fvec_t * v); |
---|
[5b41ef9] | 224 | |
---|
[83963b3] | 225 | /** compute sound pressure level (SPL) in dB |
---|
| 226 | |
---|
| 227 | This quantity is often wrongly called 'loudness'. |
---|
| 228 | |
---|
[5b41ef9] | 229 | This gives ten times the log10 of the average of the square amplitudes. |
---|
| 230 | |
---|
[83963b3] | 231 | \param v vector to compute dB SPL from |
---|
| 232 | |
---|
| 233 | \return level of v in dB SPL |
---|
| 234 | |
---|
| 235 | */ |
---|
[e9bd022] | 236 | smpl_t aubio_db_spl (const fvec_t * v); |
---|
[83963b3] | 237 | |
---|
| 238 | /** check if buffer level in dB SPL is under a given threshold |
---|
[0683ee2] | 239 | |
---|
[83963b3] | 240 | \param v vector to get level from |
---|
| 241 | \param threshold threshold in dB SPL |
---|
| 242 | |
---|
| 243 | \return 0 if level is under the given threshold, 1 otherwise |
---|
| 244 | |
---|
| 245 | */ |
---|
[e9bd022] | 246 | uint_t aubio_silence_detection (const fvec_t * v, smpl_t threshold); |
---|
[83963b3] | 247 | |
---|
| 248 | /** get buffer level if level >= threshold, 1. otherwise |
---|
| 249 | |
---|
| 250 | \param v vector to get level from |
---|
| 251 | \param threshold threshold in dB SPL |
---|
| 252 | |
---|
| 253 | \return level in dB SPL if level >= threshold, 1. otherwise |
---|
| 254 | |
---|
| 255 | */ |
---|
[24e9b0a] | 256 | smpl_t aubio_level_detection (const fvec_t * v, smpl_t threshold); |
---|
[83963b3] | 257 | |
---|
[b799241] | 258 | /** clamp the values of a vector within the range [-abs(max), abs(max)] |
---|
| 259 | |
---|
| 260 | \param in vector to clamp |
---|
[1136442] | 261 | \param absmax maximum value over which input vector elements should be clamped |
---|
[b799241] | 262 | |
---|
| 263 | */ |
---|
| 264 | void fvec_clamp(fvec_t *in, smpl_t absmax); |
---|
| 265 | |
---|
[83963b3] | 266 | #ifdef __cplusplus |
---|
| 267 | } |
---|
| 268 | #endif |
---|
| 269 | |
---|
[6f42c16] | 270 | #endif /* AUBIO_MUSICUTILS_H */ |
---|