cannam/constant-q-cpp

A C++ library and Vamp plugin implementing the Constant-Q transform of a time-domain signal.

constant-q-cpp's Issues

Type Error in FFT.cpp

ri is defined to be pointer to a double, but all of the kiss_fft function definitions take it as a float. Incompatible type error.

How to match python librosa.cqt with constant-q-cpp ?

Hi,
Could anyone get the same output than python librosa.cqt [1] with constant-q-cpp ?
constant-q-cpp has not hop_length parameter (time length of each output frame), we can only get output frames of 1/max_freq seconds.
tried to subsample the output (by averaging chunks of constant-q-cpp output) to control hop_length but I could not match the output of librosa.cqt
I also have another problem the first ~10 frames of constant-q-cpp almost equal to 0..
Any ideas ?

Here is the output I got so far, 8 octaves, 1bin per octave:
librosa.cqt(raw_audio, sr=sr, hop_length=hop_length, sparsity=0.01,
n_bins=8, bins_per_octave=1, scale=False, tuning=0)

81.77066 63.23853 85.73126 54.70133 22.289486 9.918793 2.5520785 0.21647891
75.88521 10.971143 131.40614 119.55048 40.561794 10.076316 2.4180284 0.6705376
22.214264 32.154934 235.28644 101.087166 33.556137 8.909314 2.2425086 0.6767296
23.478954 91.23524 263.20062 94.04591 27.230015 6.109539 1.6830908 0.602804
42.091568 295.96326 221.71324 40.985374 10.205679 4.3074627 2.4045377 1.4617078
42.226242 393.6818 205.13501 33.272266 9.2961855 4.230702 2.3848495 1.4510415
45.15438 487.38977 187.33626 52.973164 16.17459 5.936316 3.0049343 1.7815813
109.036606 520.87024 191.54544 48.494488 11.973509 3.007201 0.90094954 0.3631698
138.63243 485.88278 150.0018 38.036587 12.609086 5.1494775 2.7164395 1.6264207
229.58588 445.47095 128.89624 30.674294 9.998731 4.100301 2.1741393 1.302023
292.448 368.05667 99.17351 25.108074 6.6255517 1.910079 0.76821685 0.41326073
318.18692 314.9988 93.26109 23.693556 5.9957685 1.6768246 0.6439543 0.34057018
331.78436 264.88562 76.763535 19.671713 5.458373 1.8874139 0.91558653 0.5366652
305.27573 220.25504 61.272984 15.607285 4.908014 1.93713 1.0098768 0.60359365
291.54865 191.25163 55.126762 15.920821 5.224085 2.0496244 1.0622324 0.6337774
255.43016 156.81123 42.14081 11.137957 3.7328267 1.5303522 0.8096328 0.48542807
230.3502 136.72586 39.247414 11.453258 3.8305478 1.5232764 0.792822 0.47343737
200.73125 113.713486 30.808802 7.9827642 2.616217 1.0624368 0.560582 0.335974
169.69289 96.33693 27.53662 7.563389 2.347814 0.8993528 0.4637374 0.2765852
150.99515 82.90595 23.131922 5.8182926 1.812691 0.70740575 0.36589298 0.21819288
120.35955 66.39276 18.674355 4.630106 1.2619292 0.42355987 0.20073956 0.11685501
107.28719 58.5859 16.722078 4.256605 1.2394047 0.44282338 0.21800777 0.12824896
80.04482 44.47305 12.237413 2.889889 0.7032634 0.17434457 0.04924769 0.017795486

constant-q-cpp:
0 0 0 0 0 0 0 9.98623e-23
0 0 0 0 0 0 0 2.07275e-07
0 0 0 0 0 0 0 2.18631e-05
0 0 0 0 0 0 0 3.93869e-05
0 0 0 0 0 0 0 2.89571e-05
0 0 0 0 0 0 0 0.000100696
0 0 0 0 0 0 6.53269e-24 0.000299053
0 0 0 0 0 0 7.24039e-06 0.000537467
0 0 0 0 0 0 8.80853e-05 0.000867644
0 0 0 0 0 2.68308e-23 0.000262683 0.00122528
0 0 0 0 0 8.19945e-06 0.000547756 0.00157875
0 0 0 0 2.52329e-05 4.54567e-05 0.000907054 0.00183578
0.0155819 0.0153509 0.0148963 0.0141998 0.0125493 0.00971738 0.00790392 0.00621637
0.101505 0.101767 0.102588 0.104697 0.108255 0.10969 0.0781977 0.0186027
0.105382 0.105429 0.105727 0.106951 0.109555 0.111151 0.0986518 0.00984688
0.10238 0.102823 0.103907 0.106795 0.112937 0.120209 0.112378 0.0312081
0.107341 0.107281 0.107173 0.107203 0.10796 0.111972 0.116167 0.0599345
0.104164 0.104446 0.10508 0.106838 0.110942 0.116616 0.118102 0.082994
0.115768 0.115645 0.115318 0.114702 0.11342 0.111522 0.111123 0.102609
0.0909029 0.0911909 0.0918252 0.0935114 0.0979519 0.107182 0.117864 0.107549
0.116759 0.116644 0.116337 0.11578 0.114749 0.113639 0.116375 0.111216
0.111378 0.111448 0.111535 0.111845 0.112808 0.114785 0.115627 0.113197
0.105599 0.105723 0.105923 0.106471 0.108031 0.111899 0.118487 0.116803

Thanks for your help!
[1] http://librosa.org/doc/main/generated/librosa.cqt.html

Audacity 2.1.2 does not recognise the Vamp plugin

I am using Ubuntu 14.04.5 with Audacity 2.1.2 (fresh install from PPA). I have tried both the pre-compiled 1.1 (64bit) from the website, and compiling it myself. I have placed the cqvamp.so, cqvamp.n3 and cqvamp.cat file in ~/.vamp. I have also tried ~/vamp and /usr/local/Iib/vamp. I have also tried setting VAMP_PATH.

In all cases, Audacity does not show the plugin in the Add / Remove Plugins dialog. However when opening audacity with strace -eopen I see that the file cqvamp.so is opened three times when the dialog is opened, so there's some detection of the .so file. Unfortunately Audacity shows no relevant messages in dialogs, or on stderr or stdout, so I'm at a loss to know why it isn't recognising the plugin beyond opening the .so.