I’ve made some good progress on a biquad filter bank on this branch. I’ve been able to replicate Figure 10 from this reference:
That’s a filter bank with 4 individual biquad filters, here’s the C++ code that produced the plot:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 |
#include <Nsound/NsoundAll.h> #include <Nsound/biquad/FilterBank.hpp> #include <iostream> using std::cout; using namespace Nsound; using namespace Nsound::biquad; int main(void) { float64 sr = 48000; // sample rate auto khz = 1000.0; Biquad bq0(sr, 0*khz, 1*khz, 9, 6, 0, 4); Biquad bq1(sr, 4*khz, 2*khz, 12, 9, 0, 4); Biquad bq2(sr, 9*khz, 2*khz, -6, -3, 0, 4); Biquad bq3(sr, sr/2, 8*khz, 6, 3, 0, 4); FilterBank fb(sr); fb.add(bq0); fb.add(bq1); fb.add(bq2); fb.add(bq3); cout << "-----------------------------------------------------------\n" << "FilterBank: fb.to_json()\n" << "-----------------------------------------------------------\n" << fb.to_json() << "\n"; fb.plot(); Plotter pylab; pylab.ylim(-7, 13); pylab.title("N=4, Butterworth"); Plotter::show(); return 0; } |
The program also produces this output:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 |
----------------------------------------------------------- FilterBank: fb.to_json() ----------------------------------------------------------- [ { "band_width_hz": 1000, "freq_center_hz": 0, "gain_db_at_band_width": 6, "gain_db_at_fc": 9, "gain_db_baseline": 0, "order": 4, "samplerate": 48000 }, { "band_width_hz": 2000, "freq_center_hz": 4000, "gain_db_at_band_width": 9, "gain_db_at_fc": 12, "gain_db_baseline": 0, "order": 4, "samplerate": 48000 }, { "band_width_hz": 2000, "freq_center_hz": 9000, "gain_db_at_band_width": -3, "gain_db_at_fc": -6, "gain_db_baseline": 0, "order": 4, "samplerate": 48000 }, { "band_width_hz": 8000, "freq_center_hz": 24000, "gain_db_at_band_width": 3, "gain_db_at_fc": 6, "gain_db_baseline": 0, "order": 4, "samplerate": 48000 } ] |
This work is to enable a parametric equalizer application, where an audio sample can be loaded and filtered in real-time while the user adjusts the filter parameters.
I’ve been using the picojson library and it was been really easy to use.