videolan
/
ffmpeg
mirror of https://git.videolan.org/git/ffmpeg.git
1
Fork 0
Code Releases Activity

avfilter/af_asubboost: add boost option

This commit is contained in:
Paul B Mahol 2022-05-13 14:19:43 +02:00
parent 2c77d9150d
commit 075c2308e3
2 changed files with 20 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
doc/filters.texi
View File

@ -3131,15 +3131,18 @@ The filter accepts the following options:
@table @option
@item dry
Set dry gain, how much of original signal is kept. Allowed range is from 0 to 1.
Default value is 0.7.
Default value is 1.0.
@item wet
Set wet gain, how much of filtered signal is kept. Allowed range is from 0 to 1.
Default value is 0.7.
Default value is 1.0.
@item boost
Set max boost factor. Allowed range is from 1 to 12. Default value is 2.
@item decay
Set delay line decay gain value. Allowed range is from 0 to 1.
Default value is 0.7.
Default value is 0.0.
@item feedback
Set delay line feedback gain value. Allowed range is from 0 to 1.

20
libavfilter/af_asubboost.c
View File

@ -29,6 +29,7 @@ typedef struct ASubBoostContext {
double dry_gain;
double wet_gain;
double feedback;
double max_boost;
double decay;
double delay;
double cutoff;
@ -75,7 +76,7 @@ static int config_input(AVFilterLink *inlink)
ASubBoostContext *s = ctx->priv;
s->buffer = ff_get_audio_buffer(inlink, inlink->sample_rate / 10);
s->w = ff_get_audio_buffer(inlink, 2);
s->w = ff_get_audio_buffer(inlink, 3);
s->write_pos = av_calloc(inlink->ch_layout.nb_channels, sizeof(*s->write_pos));
if (!s->buffer || !s->w || !s->write_pos)
return AVERROR(ENOMEM);
@ -97,6 +98,7 @@ static int filter_channels(AVFilterContext *ctx, void *arg, int jobnr, int nb_jo
const double wet = ctx->is_disabled ? 1. : s->wet_gain;
const double dry = ctx->is_disabled ? 1. : s->dry_gain;
const double feedback = s->feedback, decay = s->decay;
const double max_boost = s->max_boost;
const double b0 = s->b0;
const double b1 = s->b1;
const double b2 = s->b2;
@ -112,16 +114,21 @@ static int filter_channels(AVFilterContext *ctx, void *arg, int jobnr, int nb_jo
double *buffer = (double *)s->buffer->extended_data[ch];
double *w = (double *)s->w->extended_data[ch];
int write_pos = s->write_pos[ch];
const double a = 0.00001;
const double b = 1. - a;
for (int n = 0; n < in->nb_samples; n++) {
double out_sample;
double out_sample, boost;
out_sample = src[n] * b0 + w[0];
w[0] = b1 * src[n] + w[1] + a1 * out_sample;
w[1] = b2 * src[n] + a2 * out_sample;
buffer[write_pos] = buffer[write_pos] * decay + out_sample * feedback;
dst[n] = (src[n] * dry + buffer[write_pos] * mix) * wet;
boost = av_clipd((1. - (fabs(src[n] * dry))) / fabs(buffer[write_pos]), 0., max_boost);
w[2] = boost > w[2] ? w[2] * b + a * boost : w[2] * a + b * boost;
w[2] = av_clipd(w[2], 0., max_boost);
dst[n] = (src[n] * dry + w[2] * buffer[write_pos] * mix) * wet;
if (++write_pos >= buffer_samples)
write_pos = 0;
@ -185,9 +192,10 @@ static int process_command(AVFilterContext *ctx, const char *cmd, const char *ar
#define FLAGS AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_RUNTIME_PARAM
static const AVOption asubboost_options[] = {
{ "dry", "set dry gain", OFFSET(dry_gain), AV_OPT_TYPE_DOUBLE, {.dbl=0.7}, 0, 1, FLAGS },
{ "wet", "set wet gain", OFFSET(wet_gain), AV_OPT_TYPE_DOUBLE, {.dbl=0.7}, 0, 1, FLAGS },
{ "decay", "set decay", OFFSET(decay), AV_OPT_TYPE_DOUBLE, {.dbl=0.7}, 0, 1, FLAGS },
{ "dry", "set dry gain", OFFSET(dry_gain), AV_OPT_TYPE_DOUBLE, {.dbl=1.0}, 0, 1, FLAGS },
{ "wet", "set wet gain", OFFSET(wet_gain), AV_OPT_TYPE_DOUBLE, {.dbl=1.0}, 0, 1, FLAGS },
{ "boost", "set max boost",OFFSET(max_boost),AV_OPT_TYPE_DOUBLE, {.dbl=2.0}, 1, 12, FLAGS },
{ "decay", "set decay", OFFSET(decay), AV_OPT_TYPE_DOUBLE, {.dbl=0.0}, 0, 1, FLAGS },
{ "feedback", "set feedback", OFFSET(feedback), AV_OPT_TYPE_DOUBLE, {.dbl=0.9}, 0, 1, FLAGS },
{ "cutoff", "set cutoff", OFFSET(cutoff), AV_OPT_TYPE_DOUBLE, {.dbl=100}, 50, 900, FLAGS },
{ "slope", "set slope", OFFSET(slope), AV_OPT_TYPE_DOUBLE, {.dbl=0.5}, 0.0001, 1, FLAGS },