1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-28 20:53:54 +02:00
FFmpeg/libavfilter/af_superequalizer.c

389 lines
11 KiB
C
Raw Normal View History

/*
* Copyright (c) 2002 Naoki Shibata
* Copyright (c) 2017 Paul B Mahol
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "libavutil/opt.h"
#include "libavcodec/avfft.h"
#include "audio.h"
#include "avfilter.h"
#include "filters.h"
#include "internal.h"
#define NBANDS 17
#define M 15
typedef struct EqParameter {
float lower, upper, gain;
} EqParameter;
typedef struct SuperEqualizerContext {
const AVClass *class;
EqParameter params[NBANDS + 1];
float gains[NBANDS + 1];
float fact[M + 1];
float aa;
float iza;
float *ires, *irest;
float *fsamples;
int winlen, tabsize;
AVFrame *in, *out;
RDFTContext *rdft, *irdft;
} SuperEqualizerContext;
static const float bands[] = {
65.406392, 92.498606, 130.81278, 184.99721, 261.62557, 369.99442, 523.25113, 739.9884, 1046.5023,
1479.9768, 2093.0045, 2959.9536, 4186.0091, 5919.9072, 8372.0181, 11839.814, 16744.036
};
static float izero(SuperEqualizerContext *s, float x)
{
float ret = 1;
int m;
for (m = 1; m <= M; m++) {
float t;
t = pow(x / 2, m) / s->fact[m];
ret += t*t;
}
return ret;
}
static float hn_lpf(int n, float f, float fs)
{
float t = 1 / fs;
float omega = 2 * M_PI * f;
if (n * omega * t == 0)
return 2 * f * t;
return 2 * f * t * sinf(n * omega * t) / (n * omega * t);
}
static float hn_imp(int n)
{
return n == 0 ? 1.f : 0.f;
}
static float hn(int n, EqParameter *param, float fs)
{
float ret, lhn;
int i;
lhn = hn_lpf(n, param[0].upper, fs);
ret = param[0].gain*lhn;
for (i = 1; i < NBANDS + 1 && param[i].upper < fs / 2; i++) {
float lhn2 = hn_lpf(n, param[i].upper, fs);
ret += param[i].gain * (lhn2 - lhn);
lhn = lhn2;
}
ret += param[i].gain * (hn_imp(n) - lhn);
return ret;
}
static float alpha(float a)
{
if (a <= 21)
return 0;
if (a <= 50)
return .5842f * pow(a - 21, 0.4f) + 0.07886f * (a - 21);
return .1102f * (a - 8.7f);
}
static float win(SuperEqualizerContext *s, float n, int N)
{
return izero(s, alpha(s->aa) * sqrtf(1 - 4 * n * n / ((N - 1) * (N - 1)))) / s->iza;
}
static void process_param(float *bc, EqParameter *param, float fs)
{
int i;
for (i = 0; i <= NBANDS; i++) {
param[i].lower = i == 0 ? 0 : bands[i - 1];
param[i].upper = i == NBANDS ? fs : bands[i];
param[i].gain = bc[i];
}
}
static int equ_init(SuperEqualizerContext *s, int wb)
{
int i,j;
s->rdft = av_rdft_init(wb, DFT_R2C);
s->irdft = av_rdft_init(wb, IDFT_C2R);
if (!s->rdft || !s->irdft)
return AVERROR(ENOMEM);
s->aa = 96;
s->winlen = (1 << (wb-1))-1;
s->tabsize = 1 << wb;
s->ires = av_calloc(s->tabsize, sizeof(float));
s->irest = av_calloc(s->tabsize, sizeof(float));
s->fsamples = av_calloc(s->tabsize, sizeof(float));
for (i = 0; i <= M; i++) {
s->fact[i] = 1;
for (j = 1; j <= i; j++)
s->fact[i] *= j;
}
s->iza = izero(s, alpha(s->aa));
return 0;
}
static void make_fir(SuperEqualizerContext *s, float *lbc, float *rbc, EqParameter *param, float fs)
{
const int winlen = s->winlen;
const int tabsize = s->tabsize;
float *nires;
int i;
if (fs <= 0)
return;
process_param(lbc, param, fs);
for (i = 0; i < winlen; i++)
s->irest[i] = hn(i - winlen / 2, param, fs) * win(s, i - winlen / 2, winlen);
for (; i < tabsize; i++)
s->irest[i] = 0;
av_rdft_calc(s->rdft, s->irest);
nires = s->ires;
for (i = 0; i < tabsize; i++)
nires[i] = s->irest[i];
}
static int filter_frame(AVFilterLink *inlink, AVFrame *in)
{
AVFilterContext *ctx = inlink->dst;
SuperEqualizerContext *s = ctx->priv;
AVFilterLink *outlink = ctx->outputs[0];
const float *ires = s->ires;
float *fsamples = s->fsamples;
int ch, i;
AVFrame *out = ff_get_audio_buffer(outlink, s->winlen);
float *src, *dst, *ptr;
if (!out) {
av_frame_free(&in);
return AVERROR(ENOMEM);
}
for (ch = 0; ch < in->channels; ch++) {
ptr = (float *)out->extended_data[ch];
dst = (float *)s->out->extended_data[ch];
src = (float *)in->extended_data[ch];
for (i = 0; i < in->nb_samples; i++)
fsamples[i] = src[i];
for (; i < s->tabsize; i++)
fsamples[i] = 0;
av_rdft_calc(s->rdft, fsamples);
fsamples[0] = ires[0] * fsamples[0];
fsamples[1] = ires[1] * fsamples[1];
for (i = 1; i < s->tabsize / 2; i++) {
float re, im;
re = ires[i*2 ] * fsamples[i*2] - ires[i*2+1] * fsamples[i*2+1];
im = ires[i*2+1] * fsamples[i*2] + ires[i*2 ] * fsamples[i*2+1];
fsamples[i*2 ] = re;
fsamples[i*2+1] = im;
}
av_rdft_calc(s->irdft, fsamples);
for (i = 0; i < s->winlen; i++)
dst[i] += fsamples[i] / s->tabsize * 2;
for (i = s->winlen; i < s->tabsize; i++)
dst[i] = fsamples[i] / s->tabsize * 2;
for (i = 0; i < s->winlen; i++)
ptr[i] = dst[i];
for (i = 0; i < s->winlen; i++)
dst[i] = dst[i+s->winlen];
}
out->pts = in->pts;
av_frame_free(&in);
return ff_filter_frame(outlink, out);
}
static int activate(AVFilterContext *ctx)
{
AVFilterLink *inlink = ctx->inputs[0];
AVFilterLink *outlink = ctx->outputs[0];
SuperEqualizerContext *s = ctx->priv;
AVFrame *in = NULL;
int ret;
FF_FILTER_FORWARD_STATUS_BACK(outlink, inlink);
ret = ff_inlink_consume_samples(inlink, s->winlen, s->winlen, &in);
if (ret < 0)
return ret;
if (ret > 0)
return filter_frame(inlink, in);
FF_FILTER_FORWARD_STATUS(inlink, outlink);
FF_FILTER_FORWARD_WANTED(outlink, inlink);
return FFERROR_NOT_READY;
}
static av_cold int init(AVFilterContext *ctx)
{
SuperEqualizerContext *s = ctx->priv;
return equ_init(s, 14);
}
static int query_formats(AVFilterContext *ctx)
{
AVFilterFormats *formats;
AVFilterChannelLayouts *layouts;
static const enum AVSampleFormat sample_fmts[] = {
AV_SAMPLE_FMT_FLTP,
AV_SAMPLE_FMT_NONE
};
int ret;
layouts = ff_all_channel_counts();
if (!layouts)
return AVERROR(ENOMEM);
ret = ff_set_common_channel_layouts(ctx, layouts);
if (ret < 0)
return ret;
formats = ff_make_format_list(sample_fmts);
if ((ret = ff_set_common_formats(ctx, formats)) < 0)
return ret;
formats = ff_all_samplerates();
return ff_set_common_samplerates(ctx, formats);
}
static int config_input(AVFilterLink *inlink)
{
AVFilterContext *ctx = inlink->dst;
SuperEqualizerContext *s = ctx->priv;
s->out = ff_get_audio_buffer(inlink, s->tabsize);
if (!s->out)
return AVERROR(ENOMEM);
return 0;
}
static int config_output(AVFilterLink *outlink)
{
AVFilterContext *ctx = outlink->src;
SuperEqualizerContext *s = ctx->priv;
make_fir(s, s->gains, s->gains, s->params, outlink->sample_rate);
return 0;
}
static av_cold void uninit(AVFilterContext *ctx)
{
SuperEqualizerContext *s = ctx->priv;
av_frame_free(&s->out);
av_freep(&s->irest);
av_freep(&s->ires);
av_freep(&s->fsamples);
av_rdft_end(s->rdft);
av_rdft_end(s->irdft);
}
static const AVFilterPad superequalizer_inputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_AUDIO,
.config_props = config_input,
},
{ NULL }
};
static const AVFilterPad superequalizer_outputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_AUDIO,
.config_props = config_output,
},
{ NULL }
};
#define AF AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
#define OFFSET(x) offsetof(SuperEqualizerContext, x)
static const AVOption superequalizer_options[] = {
{ "1b", "set 65Hz band gain", OFFSET(gains [0]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "2b", "set 92Hz band gain", OFFSET(gains [1]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "3b", "set 131Hz band gain", OFFSET(gains [2]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "4b", "set 185Hz band gain", OFFSET(gains [3]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "5b", "set 262Hz band gain", OFFSET(gains [4]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "6b", "set 370Hz band gain", OFFSET(gains [5]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "7b", "set 523Hz band gain", OFFSET(gains [6]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "8b", "set 740Hz band gain", OFFSET(gains [7]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "9b", "set 1047Hz band gain", OFFSET(gains [8]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "10b", "set 1480Hz band gain", OFFSET(gains [9]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "11b", "set 2093Hz band gain", OFFSET(gains[10]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "12b", "set 2960Hz band gain", OFFSET(gains[11]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "13b", "set 4186Hz band gain", OFFSET(gains[12]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "14b", "set 5920Hz band gain", OFFSET(gains[13]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "15b", "set 8372Hz band gain", OFFSET(gains[14]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "16b", "set 11840Hz band gain", OFFSET(gains[15]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "17b", "set 16744Hz band gain", OFFSET(gains[16]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ "18b", "set 20000Hz band gain", OFFSET(gains[17]), AV_OPT_TYPE_FLOAT, {.dbl=1}, 0, 20, AF },
{ NULL }
};
AVFILTER_DEFINE_CLASS(superequalizer);
const AVFilter ff_af_superequalizer = {
.name = "superequalizer",
.description = NULL_IF_CONFIG_SMALL("Apply 18 band equalization filter."),
.priv_size = sizeof(SuperEqualizerContext),
.priv_class = &superequalizer_class,
.query_formats = query_formats,
.init = init,
.activate = activate,
.uninit = uninit,
.inputs = superequalizer_inputs,
.outputs = superequalizer_outputs,
};