Files: voc.h, voc.c

A vocal tract physical model.
Based on the Pink Trombone algorithm by Neil Thapen, Voc implements a physical model of the vocal tract glottal pulse wave. The tract model is based on the classic Kelly-Lochbaum segmented cylindrical 1d waveguide model, and the glottal pulse wave is a LF glottal pulse model.
The soundpipe source code for Voc is generated via ctangle, one half of the literate documentation system known CWEB. The CWEB are maintained in a separate repository. They are hosted on github here: http://www.github.com/paulbatchelor/voc
This documentation is a stub. For a full overview on proper usage, consult the "Top-level functions" section of the documented code, a copy of which can be found at the Voc project page pbat.ch/proj/voc, or generate the PDF from the github page described above.


sp_voc_create(sp_voc **voc)
sp_voc_init(sp_data *sp, sp_voc *voc)
sp_voc_compute(sp_data *sp, sp_voc *voc, SPFLOAT *out)
sp_voc_destroy(sp_voc **voc)


out: Voc output.

Example Code

#include <stdlib.h>
#include <stdio.h>
#include <time.h>
#include "soundpipe.h"

typedef struct {
    sp_voc *voc;
    sp_osc *osc;
    sp_ftbl *ft;
} UserData;

void process(sp_data *sp, void *udata) {
    UserData *ud = udata;
    SPFLOAT osc = 0, voc = 0;
    sp_osc_compute(sp, ud->osc, NULL, &osc);
    if(sp_voc_get_counter(ud->voc) == 0) {
        osc = 12 + 16 * (0.5 * (osc + 1));
        sp_voc_set_tongue_shape(ud->voc, osc, 2.9);
    sp_voc_compute(sp, ud->voc, &voc);
    sp->out[0] = voc;

int main() {
    UserData ud;
    sp_data *sp;
    sp_srand(sp, 1234567);

    sp_ftbl_create(sp, &ud.ft, 2048);

    sp_voc_init(sp, ud.voc);
    sp_gen_sine(sp, ud.ft);
    sp_osc_init(sp, ud.osc, ud.ft, 0);
    ud.osc->amp = 1;
    ud.osc->freq = 0.1;

    sp->len = 44100 * 5;
    sp_process(sp, &ud, process);


    return 0;