Go to the documentation of this file.
86 double sigmae,
double *detection,
88 const double *
src,
double *dst);
91 #define OFFSET(x) offsetof(AudioDeclickContext, x)
92 #define AF AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
152 s->window_size =
inlink->sample_rate *
s->w / 1000.;
153 if (
s->window_size < 100)
155 s->ar_order =
FFMAX(
s->window_size *
s->ar / 100., 1);
156 s->nb_burst_samples =
s->window_size *
s->burst / 1000.;
157 s->hop_size =
s->window_size * (1. - (
s->overlap / 100.));
161 s->window_func_lut =
av_calloc(
s->window_size,
sizeof(*
s->window_func_lut));
162 if (!
s->window_func_lut)
164 for (
i = 0;
i <
s->window_size;
i++)
165 s->window_func_lut[
i] = sin(
M_PI *
i /
s->window_size) *
166 (1. - (
s->overlap / 100.)) *
M_PI_2;
177 if (!
s->in || !
s->out || !
s->buffer || !
s->is || !
s->enabled)
186 s->overlap_skip =
s->method ? (
s->window_size -
s->hop_size) / 2 : 0;
187 if (
s->overlap_skip > 0) {
192 s->nb_channels =
inlink->channels;
200 c->detection =
av_calloc(
s->window_size,
sizeof(*
c->detection));
201 c->auxiliary =
av_calloc(
s->ar_order + 1,
sizeof(*
c->auxiliary));
202 c->acoefficients =
av_calloc(
s->ar_order + 1,
sizeof(*
c->acoefficients));
203 c->acorrelation =
av_calloc(
s->ar_order + 1,
sizeof(*
c->acorrelation));
205 c->click =
av_calloc(
s->window_size,
sizeof(*
c->click));
206 c->index =
av_calloc(
s->window_size,
sizeof(*
c->index));
207 c->interpolated =
av_calloc(
s->window_size,
sizeof(*
c->interpolated));
208 if (!
c->auxiliary || !
c->acoefficients || !
c->detection || !
c->click ||
209 !
c->index || !
c->interpolated || !
c->acorrelation || !
c->tmp)
217 double *
output,
double scale)
221 for (
i = 0;
i <= order;
i++) {
224 for (j =
i; j <
size; j++)
242 k[0] =
a[0] = -
r[1] /
r[0];
243 alpha =
r[0] * (1. - k[0] * k[0]);
247 for (j = 0; j <
i; j++)
248 epsilon +=
a[j] *
r[
i - j];
253 for (j =
i - 1; j >= 0; j--)
254 k[j] =
a[j] + k[
i] *
a[
i - j - 1];
255 for (j = 0; j <=
i; j++)
287 while (start <= end) {
288 i = (end + start) / 2;
304 for (
i = 0;
i < n;
i++) {
305 const int in =
i * n;
309 for (j = 0; j <
i; j++)
310 value -= matrix[j * n + j] * matrix[
in + j] * matrix[
in + j];
317 for (j =
i + 1; j < n; j++) {
318 const int jn = j * n;
322 for (k = 0; k <
i; k++)
323 x -= matrix[k * n + k] * matrix[
in + k] * matrix[jn + k];
324 matrix[jn +
i] = x / matrix[
in +
i];
332 double *vector,
int n,
double *
out)
346 for (
i = 0;
i < n;
i++) {
347 const int in =
i * n;
351 for (j = 0; j <
i; j++)
352 value -= matrix[
in + j] * y[j];
356 for (
i = n - 1;
i >= 0;
i--) {
357 out[
i] = y[
i] / matrix[
i * n +
i];
358 for (j =
i + 1; j < n; j++)
359 out[
i] -= matrix[j * n +
i] *
out[j];
366 double *acoefficients,
int *
index,
int nb_errors,
367 double *auxiliary,
double *interpolated)
369 double *vector, *matrix;
372 av_fast_malloc(&
c->matrix, &
c->matrix_size, nb_errors * nb_errors *
sizeof(*
c->matrix));
384 for (
i = 0;
i < nb_errors;
i++) {
385 const int im =
i * nb_errors;
387 for (j =
i; j < nb_errors; j++) {
389 matrix[j * nb_errors +
i] = matrix[
im + j] = auxiliary[
abs(
index[j] -
index[
i])];
391 matrix[j * nb_errors +
i] = matrix[
im + j] = 0;
396 for (
i = 0;
i < nb_errors;
i++) {
411 double *unused1,
double *unused2,
413 const double *
src,
double *dst)
416 double max_amplitude = 0;
420 av_fast_malloc(&
c->histogram, &
c->histogram_size,
s->nb_hbins *
sizeof(*
c->histogram));
423 histogram =
c->histogram;
424 memset(histogram, 0,
sizeof(*histogram) *
s->nb_hbins);
426 for (
i = 0;
i <
s->window_size;
i++) {
434 for (
i =
s->nb_hbins - 1;
i > 1;
i--) {
437 max_amplitude =
i / (double)
s->nb_hbins;
443 if (max_amplitude > 0.) {
444 for (
i = 0;
i <
s->window_size;
i++) {
449 memset(
clip, 0,
s->ar_order *
sizeof(*
clip));
450 memset(
clip + (
s->window_size -
s->ar_order), 0,
s->ar_order *
sizeof(*
clip));
461 double *detection,
double *acoefficients,
463 const double *
src,
double *dst)
466 int i, j, nb_clicks = 0, prev = -1;
468 memset(detection, 0,
s->window_size *
sizeof(*detection));
471 for (j = 0; j <=
s->ar_order; j++) {
472 detection[
i] += acoefficients[j] *
src[
i - j];
476 for (
i = 0;
i <
s->window_size;
i++) {
481 for (
i = 0;
i <
s->window_size;
i++) {
486 for (j = prev + 1; j <
i; j++)
491 memset(click, 0,
s->ar_order *
sizeof(*click));
492 memset(click + (
s->window_size -
s->ar_order), 0,
s->ar_order *
sizeof(*click));
510 const double *
src = (
const double *)
s->in->extended_data[ch];
511 double *
is = (
double *)
s->is->extended_data[ch];
512 double *dst = (
double *)
s->out->extended_data[ch];
513 double *ptr = (
double *)
out->extended_data[ch];
514 double *buf = (
double *)
s->buffer->extended_data[ch];
515 const double *
w =
s->window_func_lut;
523 double *interpolated =
c->interpolated;
527 nb_errors =
s->detector(
s,
c, sigmae,
c->detection,
c->acoefficients,
530 double *enabled = (
double *)
s->enabled->extended_data[0];
533 nb_errors,
c->auxiliary, interpolated);
539 for (j = 0; j < nb_errors; j++) {
540 if (enabled[
index[j]]) {
541 dst[
index[j]] = interpolated[j];
547 memcpy(dst,
src,
s->window_size *
sizeof(*dst));
550 if (
s->method == 0) {
551 for (j = 0; j <
s->window_size; j++)
552 buf[j] += dst[j] *
w[j];
554 const int skip =
s->overlap_skip;
556 for (j = 0; j <
s->hop_size; j++)
557 buf[j] = dst[skip + j];
559 for (j = 0; j <
s->hop_size; j++)
562 memmove(buf, buf +
s->hop_size, (
s->window_size * 2 -
s->hop_size) *
sizeof(*buf));
563 memmove(
is,
is +
s->hop_size, (
s->window_size -
s->hop_size) *
sizeof(*
is));
564 memset(buf +
s->window_size * 2 -
s->hop_size, 0,
s->hop_size *
sizeof(*buf));
565 memset(
is +
s->window_size -
s->hop_size, 0,
s->hop_size *
sizeof(*
is));
576 int ret = 0, j, ch, detected_errors = 0;
593 for (ch = 0; ch <
s->in->channels; ch++) {
594 double *
is = (
double *)
s->is->extended_data[ch];
596 for (j = 0; j <
s->hop_size; j++) {
605 if (
s->samples_left > 0)
606 out->nb_samples =
FFMIN(
s->hop_size,
s->samples_left);
611 s->detected_errors += detected_errors;
612 s->nb_samples +=
out->nb_samples *
inlink->channels;
618 if (
s->samples_left > 0) {
619 s->samples_left -=
s->hop_size;
620 if (
s->samples_left <= 0)
645 double *e = (
double *)
s->enabled->extended_data[0];
652 for (
int i = 0;
i <
in->nb_samples;
i++)
653 e[
i] = !
ctx->is_disabled;
679 if (
s->eof &&
s->samples_left <= 0) {
694 s->is_declip = !strcmp(
ctx->filter->name,
"adeclip");
710 s->is_declip ?
"clips" :
"clicks",
s->detected_errors,
711 s->nb_samples, 100. *
s->detected_errors /
s->nb_samples);
723 for (
i = 0;
i <
s->nb_channels;
i++) {
737 c->histogram_size = 0;
770 .priv_class = &adeclick_class,
806 .priv_class = &adeclip_class,
void av_audio_fifo_free(AVAudioFifo *af)
Free an AVAudioFifo.
AVFrame * ff_get_audio_buffer(AVFilterLink *link, int nb_samples)
Request an audio samples buffer with a specific set of permissions.
A list of supported channel layouts.
static int query_formats(AVFilterContext *ctx)
they must not be accessed directly The fifo field contains the frames that are queued in the input for processing by the filter The status_in and status_out fields contains the queued status(EOF or error) of the link
static const AVFilterPad outputs[]
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
The official guide to swscale for confused that is
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
static enum AVSampleFormat sample_fmts[]
enum MovChannelLayoutTag * layouts
#define AVERROR_EOF
End of file.
filter_frame For filters that do not use the this method is called when a frame is pushed to the filter s input It can be called at any time except in a reentrant way If the input frame is enough to produce output
The exact code depends on how similar the blocks are and how related they are to the and needs to apply these operations to the correct inlink or outlink if there are several Macros are available to factor that when no extra processing is inlink
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
This structure describes decoded (raw) audio or video data.
static av_cold int init(AVFilterContext *ctx)
const char * name
Filter name.
AVFormatInternal * internal
An opaque field for libavformat internal usage.
A link between two filters.
static int activate(AVFilterContext *ctx)
#define FF_FILTER_FORWARD_STATUS_BACK(outlink, inlink)
Forward the status on an output link to an input link.
Context for an Audio FIFO Buffer.
int av_audio_fifo_drain(AVAudioFifo *af, int nb_samples)
Drain data from an AVAudioFifo.
A filter pad used for either input or output.
static void ff_outlink_set_status(AVFilterLink *link, int status, int64_t pts)
Set the status field of a link from the source filter.
int av_audio_fifo_write(AVAudioFifo *af, void **data, int nb_samples)
Write data to an AVAudioFifo.
int64_t av_rescale_q(int64_t a, AVRational bq, AVRational cq)
Rescale a 64-bit integer by 2 rational numbers.
static int detect_clicks(AudioDeclickContext *s, DeclickChannel *c, double sigmae, double *detection, double *acoefficients, uint8_t *click, int *index, const double *src, double *dst)
static av_cold void uninit(AVFilterContext *ctx)
static int config_input(AVFilterLink *inlink)
static void autocorrelation(const double *input, int order, int size, double *output, double scale)
Describe the class of an AVClass context structure.
static __device__ float fabs(float a)
int ff_inlink_consume_samples(AVFilterLink *link, unsigned min, unsigned max, AVFrame **rframe)
Take samples from the link's FIFO and update the link's stats.
Rational number (pair of numerator and denominator).
AVAudioFifo * av_audio_fifo_alloc(enum AVSampleFormat sample_fmt, int channels, int nb_samples)
Allocate an AVAudioFifo.
int(* detector)(struct AudioDeclickContext *s, DeclickChannel *c, double sigmae, double *detection, double *acoefficients, uint8_t *click, int *index, const double *src, double *dst)
int ff_inlink_acknowledge_status(AVFilterLink *link, int *rstatus, int64_t *rpts)
Test and acknowledge the change of status on the link.
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
static int filter_channel(AVFilterContext *ctx, void *arg, int ch, int nb_jobs)
static const AVOption adeclick_options[]
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
double fmin(double, double)
#define AV_NOPTS_VALUE
Undefined timestamp value.
The reader does not expect b to be semantically here and if the code is changed by maybe adding a a division or other the signedness will almost certainly be mistaken To avoid this confusion a new type was SUINT is the C unsigned type but it holds a signed int to use the same example SUINT a
FF_FILTER_FORWARD_WANTED(outlink, inlink)
int av_audio_fifo_size(AVAudioFifo *af)
Get the current number of samples in the AVAudioFifo available for reading.
and forward the test the status of outputs and forward it to the corresponding return FFERROR_NOT_READY If the filters stores internally one or a few frame for some input
static double autoregression(const double *samples, int ar_order, int nb_samples, double *k, double *r, double *a)
static int interpolation(DeclickChannel *c, const double *src, int ar_order, double *acoefficients, int *index, int nb_errors, double *auxiliary, double *interpolated)
#define AV_LOG_INFO
Standard information.
static int detect_clips(AudioDeclickContext *s, DeclickChannel *c, double unused0, double *unused1, double *unused2, uint8_t *clip, int *index, const double *src, double *dst)
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi - 0x80) *(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi - 0x80) *(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(const int16_t *) pi >> 8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t, *(const int16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t, *(const int16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(const int32_t *) pi >> 24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t, *(const int32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t, *(const int32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(const float *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(const float *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(const float *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(const double *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(const double *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(const double *) pi *(1U<< 31)))) #define SET_CONV_FUNC_GROUP(ofmt, ifmt) static void set_generic_function(AudioConvert *ac) { } void ff_audio_convert_free(AudioConvert **ac) { if(! *ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);} AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enum AVSampleFormat out_fmt, enum AVSampleFormat in_fmt, int channels, int sample_rate, int apply_map) { AudioConvert *ac;int in_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) return NULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method !=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt) > 2) { ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc) { av_free(ac);return NULL;} return ac;} in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar) { ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar ? ac->channels :1;} else if(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;else ac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);return ac;} int ff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in) { int use_generic=1;int len=in->nb_samples;int p;if(ac->dc) { av_log(ac->avr, AV_LOG_TRACE, "%d samples - audio_convert: %s to %s (dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));return ff_convert_dither(ac-> in
static int filter_frame(AVFilterLink *inlink)
static int find_index(int *index, int value, int size)
static const AVOption adeclip_options[]
AVSampleFormat
Audio sample formats.
Used for passing data between threads.
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf default value
const char * name
Pad name.
static int factorization(double *matrix, int n)
AVFILTER_DEFINE_CLASS(adeclick)
static const AVFilterPad inputs[]
void * av_calloc(size_t nmemb, size_t size)
Non-inlined equivalent of av_mallocz_array().
@ AV_SAMPLE_FMT_DBLP
double, planar
Filter the word “frame” indicates either a video frame or a group of audio samples
AVRational time_base
Define the time base used by the PTS of the frames/samples which will pass through this link.
#define AVFILTER_FLAG_SLICE_THREADS
The filter supports multithreading by splitting frames into multiple parts and processing them concur...
static const int16_t alpha[]
static int do_interpolation(DeclickChannel *c, double *matrix, double *vector, int n, double *out)
void av_fast_malloc(void *ptr, unsigned int *size, size_t min_size)
Allocate a buffer, reusing the given one if large enough.
#define AVFILTER_FLAG_SUPPORT_TIMELINE_INTERNAL
Same as AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC, except that the filter will have its filter_frame() c...
#define flags(name, subs,...)
static int isfinite_array(double *samples, int nb_samples)
int av_audio_fifo_peek(AVAudioFifo *af, void **data, int nb_samples)
Peek data from an AVAudioFifo.
static double clip(void *opaque, double val)
Clip value val in the minval - maxval range.
void ff_filter_set_ready(AVFilterContext *filter, unsigned priority)
Mark a filter ready and schedule it for activation.