FFmpeg/libavfilter/af_afade.c
Ganesh Ajjanagadde 9ee1feaa7c avfilter/af_afade: improve accuracy and speed of gain computation
Gain computation for various curves was being done in a needlessly
inaccurate fashion. Of course these are all subjective curves, but when
a curve is advertised to the user, it should be matched as closely as
possible within the limitations of libm. In particular, the constants
kept here were pretty inaccurate for double precision.

Speed improvements are mainly due to the avoidance of pow, the most
notorious of the libm functions in terms of performance. To be fair, it
is the GNU libm that is among the worst, but it is not really GNU libm's fault
since others simply yield a higher error as measured in ULP.

"Magic" constants are also accordingly documented, since they take at
least a minute of thought for a casual reader.

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
2015-11-26 09:20:46 -05:00

686 lines
33 KiB
C

/*
* Copyright (c) 2013-2015 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
*/
/**
* @file
* fade audio filter
*/
#include "libavutil/audio_fifo.h"
#include "libavutil/opt.h"
#include "audio.h"
#include "avfilter.h"
#include "internal.h"
typedef struct {
const AVClass *class;
int type;
int curve, curve2;
int nb_samples;
int64_t start_sample;
int64_t duration;
int64_t start_time;
int overlap;
int cf0_eof;
int crossfade_is_over;
AVAudioFifo *fifo[2];
int64_t pts;
void (*fade_samples)(uint8_t **dst, uint8_t * const *src,
int nb_samples, int channels, int direction,
int64_t start, int range, int curve);
void (*crossfade_samples)(uint8_t **dst, uint8_t * const *cf0,
uint8_t * const *cf1,
int nb_samples, int channels,
int curve0, int curve1);
} AudioFadeContext;
enum CurveType { TRI, QSIN, ESIN, HSIN, LOG, IPAR, QUA, CUB, SQU, CBR, PAR, EXP, IQSIN, IHSIN, DESE, DESI, NB_CURVES };
#define OFFSET(x) offsetof(AudioFadeContext, x)
#define FLAGS AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
static int query_formats(AVFilterContext *ctx)
{
AVFilterFormats *formats;
AVFilterChannelLayouts *layouts;
static const enum AVSampleFormat sample_fmts[] = {
AV_SAMPLE_FMT_S16, AV_SAMPLE_FMT_S16P,
AV_SAMPLE_FMT_S32, AV_SAMPLE_FMT_S32P,
AV_SAMPLE_FMT_FLT, AV_SAMPLE_FMT_FLTP,
AV_SAMPLE_FMT_DBL, AV_SAMPLE_FMT_DBLP,
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 (!formats)
return AVERROR(ENOMEM);
ret = ff_set_common_formats(ctx, formats);
if (ret < 0)
return ret;
formats = ff_all_samplerates();
if (!formats)
return AVERROR(ENOMEM);
return ff_set_common_samplerates(ctx, formats);
}
static double fade_gain(int curve, int64_t index, int range)
{
#define CUBE(a) ((a)*(a)*(a))
double gain;
gain = av_clipd(1.0 * index / range, 0, 1.0);
switch (curve) {
case QSIN:
gain = sin(gain * M_PI / 2.0);
break;
case IQSIN:
/* 0.6... = 2 / M_PI */
gain = 0.6366197723675814 * asin(gain);
break;
case ESIN:
gain = 1.0 - cos(M_PI / 4.0 * (CUBE(2.0*gain - 1) + 1));
break;
case HSIN:
gain = (1.0 - cos(gain * M_PI)) / 2.0;
break;
case IHSIN:
/* 0.3... = 1 / M_PI */
gain = 0.3183098861837907 * acos(1 - 2 * gain);
break;
case EXP:
/* -11.5... = 5*ln(0.1) */
gain = exp(-11.512925464970227 * (1 - gain));
break;
case LOG:
gain = av_clipd(1 + 0.2 * log10(gain), 0, 1.0);
break;
case PAR:
gain = 1 - sqrt(1 - gain);
break;
case IPAR:
gain = (1 - (1 - gain) * (1 - gain));
break;
case QUA:
gain *= gain;
break;
case CUB:
gain = CUBE(gain);
break;
case SQU:
gain = sqrt(gain);
break;
case CBR:
gain = cbrt(gain);
break;
case DESE:
gain = gain <= 0.5 ? cbrt(2 * gain) / 2: 1 - cbrt(2 * (1 - gain)) / 2;
break;
case DESI:
gain = gain <= 0.5 ? CUBE(2 * gain) / 2: 1 - CUBE(2 * (1 - gain)) / 2;
break;
}
return gain;
}
#define FADE_PLANAR(name, type) \
static void fade_samples_## name ##p(uint8_t **dst, uint8_t * const *src, \
int nb_samples, int channels, int dir, \
int64_t start, int range, int curve) \
{ \
int i, c; \
\
for (i = 0; i < nb_samples; i++) { \
double gain = fade_gain(curve, start + i * dir, range); \
for (c = 0; c < channels; c++) { \
type *d = (type *)dst[c]; \
const type *s = (type *)src[c]; \
\
d[i] = s[i] * gain; \
} \
} \
}
#define FADE(name, type) \
static void fade_samples_## name (uint8_t **dst, uint8_t * const *src, \
int nb_samples, int channels, int dir, \
int64_t start, int range, int curve) \
{ \
type *d = (type *)dst[0]; \
const type *s = (type *)src[0]; \
int i, c, k = 0; \
\
for (i = 0; i < nb_samples; i++) { \
double gain = fade_gain(curve, start + i * dir, range); \
for (c = 0; c < channels; c++, k++) \
d[k] = s[k] * gain; \
} \
}
FADE_PLANAR(dbl, double)
FADE_PLANAR(flt, float)
FADE_PLANAR(s16, int16_t)
FADE_PLANAR(s32, int32_t)
FADE(dbl, double)
FADE(flt, float)
FADE(s16, int16_t)
FADE(s32, int32_t)
static int config_output(AVFilterLink *outlink)
{
AVFilterContext *ctx = outlink->src;
AudioFadeContext *s = ctx->priv;
switch (outlink->format) {
case AV_SAMPLE_FMT_DBL: s->fade_samples = fade_samples_dbl; break;
case AV_SAMPLE_FMT_DBLP: s->fade_samples = fade_samples_dblp; break;
case AV_SAMPLE_FMT_FLT: s->fade_samples = fade_samples_flt; break;
case AV_SAMPLE_FMT_FLTP: s->fade_samples = fade_samples_fltp; break;
case AV_SAMPLE_FMT_S16: s->fade_samples = fade_samples_s16; break;
case AV_SAMPLE_FMT_S16P: s->fade_samples = fade_samples_s16p; break;
case AV_SAMPLE_FMT_S32: s->fade_samples = fade_samples_s32; break;
case AV_SAMPLE_FMT_S32P: s->fade_samples = fade_samples_s32p; break;
}
if (s->duration)
s->nb_samples = av_rescale(s->duration, outlink->sample_rate, AV_TIME_BASE);
if (s->start_time)
s->start_sample = av_rescale(s->start_time, outlink->sample_rate, AV_TIME_BASE);
return 0;
}
#if CONFIG_AFADE_FILTER
static const AVOption afade_options[] = {
{ "type", "set the fade direction", OFFSET(type), AV_OPT_TYPE_INT, {.i64 = 0 }, 0, 1, FLAGS, "type" },
{ "t", "set the fade direction", OFFSET(type), AV_OPT_TYPE_INT, {.i64 = 0 }, 0, 1, FLAGS, "type" },
{ "in", "fade-in", 0, AV_OPT_TYPE_CONST, {.i64 = 0 }, 0, 0, FLAGS, "type" },
{ "out", "fade-out", 0, AV_OPT_TYPE_CONST, {.i64 = 1 }, 0, 0, FLAGS, "type" },
{ "start_sample", "set number of first sample to start fading", OFFSET(start_sample), AV_OPT_TYPE_INT64, {.i64 = 0 }, 0, INT64_MAX, FLAGS },
{ "ss", "set number of first sample to start fading", OFFSET(start_sample), AV_OPT_TYPE_INT64, {.i64 = 0 }, 0, INT64_MAX, FLAGS },
{ "nb_samples", "set number of samples for fade duration", OFFSET(nb_samples), AV_OPT_TYPE_INT, {.i64 = 44100}, 1, INT32_MAX, FLAGS },
{ "ns", "set number of samples for fade duration", OFFSET(nb_samples), AV_OPT_TYPE_INT, {.i64 = 44100}, 1, INT32_MAX, FLAGS },
{ "start_time", "set time to start fading", OFFSET(start_time), AV_OPT_TYPE_DURATION, {.i64 = 0. }, 0, INT32_MAX, FLAGS },
{ "st", "set time to start fading", OFFSET(start_time), AV_OPT_TYPE_DURATION, {.i64 = 0. }, 0, INT32_MAX, FLAGS },
{ "duration", "set fade duration", OFFSET(duration), AV_OPT_TYPE_DURATION, {.i64 = 0. }, 0, INT32_MAX, FLAGS },
{ "d", "set fade duration", OFFSET(duration), AV_OPT_TYPE_DURATION, {.i64 = 0. }, 0, INT32_MAX, FLAGS },
{ "curve", "set fade curve type", OFFSET(curve), AV_OPT_TYPE_INT, {.i64 = TRI }, 0, NB_CURVES - 1, FLAGS, "curve" },
{ "c", "set fade curve type", OFFSET(curve), AV_OPT_TYPE_INT, {.i64 = TRI }, 0, NB_CURVES - 1, FLAGS, "curve" },
{ "tri", "linear slope", 0, AV_OPT_TYPE_CONST, {.i64 = TRI }, 0, 0, FLAGS, "curve" },
{ "qsin", "quarter of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = QSIN }, 0, 0, FLAGS, "curve" },
{ "esin", "exponential sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = ESIN }, 0, 0, FLAGS, "curve" },
{ "hsin", "half of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = HSIN }, 0, 0, FLAGS, "curve" },
{ "log", "logarithmic", 0, AV_OPT_TYPE_CONST, {.i64 = LOG }, 0, 0, FLAGS, "curve" },
{ "ipar", "inverted parabola", 0, AV_OPT_TYPE_CONST, {.i64 = IPAR }, 0, 0, FLAGS, "curve" },
{ "qua", "quadratic", 0, AV_OPT_TYPE_CONST, {.i64 = QUA }, 0, 0, FLAGS, "curve" },
{ "cub", "cubic", 0, AV_OPT_TYPE_CONST, {.i64 = CUB }, 0, 0, FLAGS, "curve" },
{ "squ", "square root", 0, AV_OPT_TYPE_CONST, {.i64 = SQU }, 0, 0, FLAGS, "curve" },
{ "cbr", "cubic root", 0, AV_OPT_TYPE_CONST, {.i64 = CBR }, 0, 0, FLAGS, "curve" },
{ "par", "parabola", 0, AV_OPT_TYPE_CONST, {.i64 = PAR }, 0, 0, FLAGS, "curve" },
{ "exp", "exponential", 0, AV_OPT_TYPE_CONST, {.i64 = EXP }, 0, 0, FLAGS, "curve" },
{ "iqsin", "inverted quarter of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = IQSIN}, 0, 0, FLAGS, "curve" },
{ "ihsin", "inverted half of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = IHSIN}, 0, 0, FLAGS, "curve" },
{ "dese", "double-exponential seat", 0, AV_OPT_TYPE_CONST, {.i64 = DESE }, 0, 0, FLAGS, "curve" },
{ "desi", "double-exponential sigmoid", 0, AV_OPT_TYPE_CONST, {.i64 = DESI }, 0, 0, FLAGS, "curve" },
{ NULL }
};
AVFILTER_DEFINE_CLASS(afade);
static av_cold int init(AVFilterContext *ctx)
{
AudioFadeContext *s = ctx->priv;
if (INT64_MAX - s->nb_samples < s->start_sample)
return AVERROR(EINVAL);
return 0;
}
static int filter_frame(AVFilterLink *inlink, AVFrame *buf)
{
AudioFadeContext *s = inlink->dst->priv;
AVFilterLink *outlink = inlink->dst->outputs[0];
int nb_samples = buf->nb_samples;
AVFrame *out_buf;
int64_t cur_sample = av_rescale_q(buf->pts, inlink->time_base, (AVRational){1, inlink->sample_rate});
if ((!s->type && (s->start_sample + s->nb_samples < cur_sample)) ||
( s->type && (cur_sample + nb_samples < s->start_sample)))
return ff_filter_frame(outlink, buf);
if (av_frame_is_writable(buf)) {
out_buf = buf;
} else {
out_buf = ff_get_audio_buffer(inlink, nb_samples);
if (!out_buf)
return AVERROR(ENOMEM);
av_frame_copy_props(out_buf, buf);
}
if ((!s->type && (cur_sample + nb_samples < s->start_sample)) ||
( s->type && (s->start_sample + s->nb_samples < cur_sample))) {
av_samples_set_silence(out_buf->extended_data, 0, nb_samples,
av_frame_get_channels(out_buf), out_buf->format);
} else {
int64_t start;
if (!s->type)
start = cur_sample - s->start_sample;
else
start = s->start_sample + s->nb_samples - cur_sample;
s->fade_samples(out_buf->extended_data, buf->extended_data,
nb_samples, av_frame_get_channels(buf),
s->type ? -1 : 1, start,
s->nb_samples, s->curve);
}
if (buf != out_buf)
av_frame_free(&buf);
return ff_filter_frame(outlink, out_buf);
}
static const AVFilterPad avfilter_af_afade_inputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_AUDIO,
.filter_frame = filter_frame,
},
{ NULL }
};
static const AVFilterPad avfilter_af_afade_outputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_AUDIO,
.config_props = config_output,
},
{ NULL }
};
AVFilter ff_af_afade = {
.name = "afade",
.description = NULL_IF_CONFIG_SMALL("Fade in/out input audio."),
.query_formats = query_formats,
.priv_size = sizeof(AudioFadeContext),
.init = init,
.inputs = avfilter_af_afade_inputs,
.outputs = avfilter_af_afade_outputs,
.priv_class = &afade_class,
.flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC,
};
#endif /* CONFIG_AFADE_FILTER */
#if CONFIG_ACROSSFADE_FILTER
static const AVOption acrossfade_options[] = {
{ "nb_samples", "set number of samples for cross fade duration", OFFSET(nb_samples), AV_OPT_TYPE_INT, {.i64 = 44100}, 1, INT32_MAX/10, FLAGS },
{ "ns", "set number of samples for cross fade duration", OFFSET(nb_samples), AV_OPT_TYPE_INT, {.i64 = 44100}, 1, INT32_MAX/10, FLAGS },
{ "duration", "set cross fade duration", OFFSET(duration), AV_OPT_TYPE_DURATION, {.i64 = 0. }, 0, 60, FLAGS },
{ "d", "set cross fade duration", OFFSET(duration), AV_OPT_TYPE_DURATION, {.i64 = 0. }, 0, 60, FLAGS },
{ "overlap", "overlap 1st stream end with 2nd stream start", OFFSET(overlap), AV_OPT_TYPE_BOOL, {.i64 = 1 }, 0, 1, FLAGS },
{ "o", "overlap 1st stream end with 2nd stream start", OFFSET(overlap), AV_OPT_TYPE_BOOL, {.i64 = 1 }, 0, 1, FLAGS },
{ "curve1", "set fade curve type for 1st stream", OFFSET(curve), AV_OPT_TYPE_INT, {.i64 = TRI }, 0, NB_CURVES - 1, FLAGS, "curve1" },
{ "c1", "set fade curve type for 1st stream", OFFSET(curve), AV_OPT_TYPE_INT, {.i64 = TRI }, 0, NB_CURVES - 1, FLAGS, "curve1" },
{ "tri", "linear slope", 0, AV_OPT_TYPE_CONST, {.i64 = TRI }, 0, 0, FLAGS, "curve1" },
{ "qsin", "quarter of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = QSIN }, 0, 0, FLAGS, "curve1" },
{ "esin", "exponential sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = ESIN }, 0, 0, FLAGS, "curve1" },
{ "hsin", "half of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = HSIN }, 0, 0, FLAGS, "curve1" },
{ "log", "logarithmic", 0, AV_OPT_TYPE_CONST, {.i64 = LOG }, 0, 0, FLAGS, "curve1" },
{ "ipar", "inverted parabola", 0, AV_OPT_TYPE_CONST, {.i64 = IPAR }, 0, 0, FLAGS, "curve1" },
{ "qua", "quadratic", 0, AV_OPT_TYPE_CONST, {.i64 = QUA }, 0, 0, FLAGS, "curve1" },
{ "cub", "cubic", 0, AV_OPT_TYPE_CONST, {.i64 = CUB }, 0, 0, FLAGS, "curve1" },
{ "squ", "square root", 0, AV_OPT_TYPE_CONST, {.i64 = SQU }, 0, 0, FLAGS, "curve1" },
{ "cbr", "cubic root", 0, AV_OPT_TYPE_CONST, {.i64 = CBR }, 0, 0, FLAGS, "curve1" },
{ "par", "parabola", 0, AV_OPT_TYPE_CONST, {.i64 = PAR }, 0, 0, FLAGS, "curve1" },
{ "exp", "exponential", 0, AV_OPT_TYPE_CONST, {.i64 = EXP }, 0, 0, FLAGS, "curve1" },
{ "iqsin", "inverted quarter of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = IQSIN}, 0, 0, FLAGS, "curve1" },
{ "ihsin", "inverted half of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = IHSIN}, 0, 0, FLAGS, "curve1" },
{ "dese", "double-exponential seat", 0, AV_OPT_TYPE_CONST, {.i64 = DESE }, 0, 0, FLAGS, "curve1" },
{ "desi", "double-exponential sigmoid", 0, AV_OPT_TYPE_CONST, {.i64 = DESI }, 0, 0, FLAGS, "curve1" },
{ "curve2", "set fade curve type for 2nd stream", OFFSET(curve2), AV_OPT_TYPE_INT, {.i64 = TRI }, 0, NB_CURVES - 1, FLAGS, "curve2" },
{ "c2", "set fade curve type for 2nd stream", OFFSET(curve2), AV_OPT_TYPE_INT, {.i64 = TRI }, 0, NB_CURVES - 1, FLAGS, "curve2" },
{ "tri", "linear slope", 0, AV_OPT_TYPE_CONST, {.i64 = TRI }, 0, 0, FLAGS, "curve2" },
{ "qsin", "quarter of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = QSIN }, 0, 0, FLAGS, "curve2" },
{ "esin", "exponential sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = ESIN }, 0, 0, FLAGS, "curve2" },
{ "hsin", "half of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = HSIN }, 0, 0, FLAGS, "curve2" },
{ "log", "logarithmic", 0, AV_OPT_TYPE_CONST, {.i64 = LOG }, 0, 0, FLAGS, "curve2" },
{ "ipar", "inverted parabola", 0, AV_OPT_TYPE_CONST, {.i64 = IPAR }, 0, 0, FLAGS, "curve2" },
{ "qua", "quadratic", 0, AV_OPT_TYPE_CONST, {.i64 = QUA }, 0, 0, FLAGS, "curve2" },
{ "cub", "cubic", 0, AV_OPT_TYPE_CONST, {.i64 = CUB }, 0, 0, FLAGS, "curve2" },
{ "squ", "square root", 0, AV_OPT_TYPE_CONST, {.i64 = SQU }, 0, 0, FLAGS, "curve2" },
{ "cbr", "cubic root", 0, AV_OPT_TYPE_CONST, {.i64 = CBR }, 0, 0, FLAGS, "curve2" },
{ "par", "parabola", 0, AV_OPT_TYPE_CONST, {.i64 = PAR }, 0, 0, FLAGS, "curve2" },
{ "exp", "exponential", 0, AV_OPT_TYPE_CONST, {.i64 = EXP }, 0, 0, FLAGS, "curve2" },
{ "iqsin", "inverted quarter of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = IQSIN}, 0, 0, FLAGS, "curve2" },
{ "ihsin", "inverted half of sine wave", 0, AV_OPT_TYPE_CONST, {.i64 = IHSIN}, 0, 0, FLAGS, "curve2" },
{ "dese", "double-exponential seat", 0, AV_OPT_TYPE_CONST, {.i64 = DESE }, 0, 0, FLAGS, "curve2" },
{ "desi", "double-exponential sigmoid", 0, AV_OPT_TYPE_CONST, {.i64 = DESI }, 0, 0, FLAGS, "curve2" },
{ NULL }
};
AVFILTER_DEFINE_CLASS(acrossfade);
#define CROSSFADE_PLANAR(name, type) \
static void crossfade_samples_## name ##p(uint8_t **dst, uint8_t * const *cf0, \
uint8_t * const *cf1, \
int nb_samples, int channels, \
int curve0, int curve1) \
{ \
int i, c; \
\
for (i = 0; i < nb_samples; i++) { \
double gain0 = fade_gain(curve0, nb_samples - 1 - i, nb_samples); \
double gain1 = fade_gain(curve1, i, nb_samples); \
for (c = 0; c < channels; c++) { \
type *d = (type *)dst[c]; \
const type *s0 = (type *)cf0[c]; \
const type *s1 = (type *)cf1[c]; \
\
d[i] = s0[i] * gain0 + s1[i] * gain1; \
} \
} \
}
#define CROSSFADE(name, type) \
static void crossfade_samples_## name (uint8_t **dst, uint8_t * const *cf0, \
uint8_t * const *cf1, \
int nb_samples, int channels, \
int curve0, int curve1) \
{ \
type *d = (type *)dst[0]; \
const type *s0 = (type *)cf0[0]; \
const type *s1 = (type *)cf1[0]; \
int i, c, k = 0; \
\
for (i = 0; i < nb_samples; i++) { \
double gain0 = fade_gain(curve0, nb_samples - 1 - i, nb_samples); \
double gain1 = fade_gain(curve1, i, nb_samples); \
for (c = 0; c < channels; c++, k++) \
d[k] = s0[k] * gain0 + s1[k] * gain1; \
} \
}
CROSSFADE_PLANAR(dbl, double)
CROSSFADE_PLANAR(flt, float)
CROSSFADE_PLANAR(s16, int16_t)
CROSSFADE_PLANAR(s32, int32_t)
CROSSFADE(dbl, double)
CROSSFADE(flt, float)
CROSSFADE(s16, int16_t)
CROSSFADE(s32, int32_t)
static int acrossfade_filter_frame(AVFilterLink *inlink, AVFrame *in)
{
AVFilterContext *ctx = inlink->dst;
AudioFadeContext *s = ctx->priv;
AVFilterLink *outlink = ctx->outputs[0];
AVFrame *out, *cf[2] = { NULL };
int ret = 0, nb_samples;
if (s->crossfade_is_over) {
in->pts = s->pts;
s->pts += av_rescale_q(in->nb_samples,
(AVRational){ 1, outlink->sample_rate }, outlink->time_base);
return ff_filter_frame(outlink, in);
} else if (inlink == ctx->inputs[0]) {
av_audio_fifo_write(s->fifo[0], (void **)in->extended_data, in->nb_samples);
nb_samples = av_audio_fifo_size(s->fifo[0]) - s->nb_samples;
if (nb_samples > 0) {
out = ff_get_audio_buffer(outlink, nb_samples);
if (!out) {
ret = AVERROR(ENOMEM);
goto fail;
}
av_audio_fifo_read(s->fifo[0], (void **)out->extended_data, nb_samples);
out->pts = s->pts;
s->pts += av_rescale_q(nb_samples,
(AVRational){ 1, outlink->sample_rate }, outlink->time_base);
ret = ff_filter_frame(outlink, out);
}
} else if (av_audio_fifo_size(s->fifo[1]) < s->nb_samples) {
if (!s->overlap && av_audio_fifo_size(s->fifo[0]) > 0) {
nb_samples = av_audio_fifo_size(s->fifo[0]);
cf[0] = ff_get_audio_buffer(outlink, nb_samples);
out = ff_get_audio_buffer(outlink, nb_samples);
if (!out || !cf[0]) {
ret = AVERROR(ENOMEM);
goto fail;
}
av_audio_fifo_read(s->fifo[0], (void **)cf[0]->extended_data, nb_samples);
s->fade_samples(out->extended_data, cf[0]->extended_data, nb_samples,
outlink->channels, -1, nb_samples - 1, nb_samples, s->curve);
out->pts = s->pts;
s->pts += av_rescale_q(nb_samples,
(AVRational){ 1, outlink->sample_rate }, outlink->time_base);
ret = ff_filter_frame(outlink, out);
if (ret < 0)
goto fail;
}
av_audio_fifo_write(s->fifo[1], (void **)in->extended_data, in->nb_samples);
} else if (av_audio_fifo_size(s->fifo[1]) >= s->nb_samples) {
av_audio_fifo_write(s->fifo[1], (void **)in->extended_data, in->nb_samples);
if (s->overlap) {
cf[0] = ff_get_audio_buffer(outlink, s->nb_samples);
cf[1] = ff_get_audio_buffer(outlink, s->nb_samples);
out = ff_get_audio_buffer(outlink, s->nb_samples);
if (!out || !cf[0] || !cf[1]) {
av_frame_free(&out);
ret = AVERROR(ENOMEM);
goto fail;
}
av_audio_fifo_read(s->fifo[0], (void **)cf[0]->extended_data, s->nb_samples);
av_audio_fifo_read(s->fifo[1], (void **)cf[1]->extended_data, s->nb_samples);
s->crossfade_samples(out->extended_data, cf[0]->extended_data,
cf[1]->extended_data,
s->nb_samples, av_frame_get_channels(in),
s->curve, s->curve2);
out->pts = s->pts;
s->pts += av_rescale_q(s->nb_samples,
(AVRational){ 1, outlink->sample_rate }, outlink->time_base);
ret = ff_filter_frame(outlink, out);
if (ret < 0)
goto fail;
} else {
out = ff_get_audio_buffer(outlink, s->nb_samples);
cf[1] = ff_get_audio_buffer(outlink, s->nb_samples);
if (!out || !cf[1]) {
ret = AVERROR(ENOMEM);
av_frame_free(&out);
goto fail;
}
av_audio_fifo_read(s->fifo[1], (void **)cf[1]->extended_data, s->nb_samples);
s->fade_samples(out->extended_data, cf[1]->extended_data, s->nb_samples,
outlink->channels, 1, 0, s->nb_samples, s->curve2);
out->pts = s->pts;
s->pts += av_rescale_q(s->nb_samples,
(AVRational){ 1, outlink->sample_rate }, outlink->time_base);
ret = ff_filter_frame(outlink, out);
if (ret < 0)
goto fail;
}
nb_samples = av_audio_fifo_size(s->fifo[1]);
if (nb_samples > 0) {
out = ff_get_audio_buffer(outlink, nb_samples);
if (!out) {
ret = AVERROR(ENOMEM);
goto fail;
}
av_audio_fifo_read(s->fifo[1], (void **)out->extended_data, nb_samples);
out->pts = s->pts;
s->pts += av_rescale_q(nb_samples,
(AVRational){ 1, outlink->sample_rate }, outlink->time_base);
ret = ff_filter_frame(outlink, out);
}
s->crossfade_is_over = 1;
}
fail:
av_frame_free(&in);
av_frame_free(&cf[0]);
av_frame_free(&cf[1]);
return ret;
}
static int acrossfade_request_frame(AVFilterLink *outlink)
{
AVFilterContext *ctx = outlink->src;
AudioFadeContext *s = ctx->priv;
int ret = 0;
if (!s->cf0_eof) {
AVFilterLink *cf0 = ctx->inputs[0];
ret = ff_request_frame(cf0);
if (ret < 0 && ret != AVERROR_EOF)
return ret;
if (ret == AVERROR_EOF) {
s->cf0_eof = 1;
ret = 0;
}
} else {
AVFilterLink *cf1 = ctx->inputs[1];
int nb_samples = av_audio_fifo_size(s->fifo[1]);
ret = ff_request_frame(cf1);
if (ret == AVERROR_EOF && nb_samples > 0) {
AVFrame *out = ff_get_audio_buffer(outlink, nb_samples);
if (!out)
return AVERROR(ENOMEM);
av_audio_fifo_read(s->fifo[1], (void **)out->extended_data, nb_samples);
ret = ff_filter_frame(outlink, out);
}
}
return ret;
}
static int acrossfade_config_output(AVFilterLink *outlink)
{
AVFilterContext *ctx = outlink->src;
AudioFadeContext *s = ctx->priv;
if (ctx->inputs[0]->sample_rate != ctx->inputs[1]->sample_rate) {
av_log(ctx, AV_LOG_ERROR,
"Inputs must have the same sample rate "
"%d for in0 vs %d for in1\n",
ctx->inputs[0]->sample_rate, ctx->inputs[1]->sample_rate);
return AVERROR(EINVAL);
}
outlink->sample_rate = ctx->inputs[0]->sample_rate;
outlink->time_base = ctx->inputs[0]->time_base;
outlink->channel_layout = ctx->inputs[0]->channel_layout;
outlink->channels = ctx->inputs[0]->channels;
switch (outlink->format) {
case AV_SAMPLE_FMT_DBL: s->crossfade_samples = crossfade_samples_dbl; break;
case AV_SAMPLE_FMT_DBLP: s->crossfade_samples = crossfade_samples_dblp; break;
case AV_SAMPLE_FMT_FLT: s->crossfade_samples = crossfade_samples_flt; break;
case AV_SAMPLE_FMT_FLTP: s->crossfade_samples = crossfade_samples_fltp; break;
case AV_SAMPLE_FMT_S16: s->crossfade_samples = crossfade_samples_s16; break;
case AV_SAMPLE_FMT_S16P: s->crossfade_samples = crossfade_samples_s16p; break;
case AV_SAMPLE_FMT_S32: s->crossfade_samples = crossfade_samples_s32; break;
case AV_SAMPLE_FMT_S32P: s->crossfade_samples = crossfade_samples_s32p; break;
}
config_output(outlink);
s->fifo[0] = av_audio_fifo_alloc(outlink->format, outlink->channels, s->nb_samples);
s->fifo[1] = av_audio_fifo_alloc(outlink->format, outlink->channels, s->nb_samples);
if (!s->fifo[0] || !s->fifo[1])
return AVERROR(ENOMEM);
return 0;
}
static av_cold void uninit(AVFilterContext *ctx)
{
AudioFadeContext *s = ctx->priv;
av_audio_fifo_free(s->fifo[0]);
av_audio_fifo_free(s->fifo[1]);
}
static const AVFilterPad avfilter_af_acrossfade_inputs[] = {
{
.name = "crossfade0",
.type = AVMEDIA_TYPE_AUDIO,
.filter_frame = acrossfade_filter_frame,
},
{
.name = "crossfade1",
.type = AVMEDIA_TYPE_AUDIO,
.filter_frame = acrossfade_filter_frame,
},
{ NULL }
};
static const AVFilterPad avfilter_af_acrossfade_outputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_AUDIO,
.request_frame = acrossfade_request_frame,
.config_props = acrossfade_config_output,
},
{ NULL }
};
AVFilter ff_af_acrossfade = {
.name = "acrossfade",
.description = NULL_IF_CONFIG_SMALL("Cross fade two input audio streams."),
.query_formats = query_formats,
.priv_size = sizeof(AudioFadeContext),
.uninit = uninit,
.priv_class = &acrossfade_class,
.inputs = avfilter_af_acrossfade_inputs,
.outputs = avfilter_af_acrossfade_outputs,
};
#endif /* CONFIG_ACROSSFADE_FILTER */