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
144 s->window_size =
inlink->sample_rate *
s->w / 1000.;
145 if (
s->window_size < 100)
147 s->ar_order =
FFMAX(
s->window_size *
s->ar / 100., 1);
148 s->nb_burst_samples =
s->window_size *
s->burst / 1000.;
149 s->hop_size =
s->window_size * (1. - (
s->overlap / 100.));
153 s->window_func_lut =
av_calloc(
s->window_size,
sizeof(*
s->window_func_lut));
154 if (!
s->window_func_lut)
156 for (
i = 0;
i <
s->window_size;
i++)
157 s->window_func_lut[
i] = sin(
M_PI *
i /
s->window_size) *
158 (1. - (
s->overlap / 100.)) *
M_PI_2;
169 if (!
s->in || !
s->out || !
s->buffer || !
s->is || !
s->enabled)
178 s->overlap_skip =
s->method ? (
s->window_size -
s->hop_size) / 2 : 0;
179 if (
s->overlap_skip > 0) {
184 s->nb_channels =
inlink->channels;
192 c->detection =
av_calloc(
s->window_size,
sizeof(*
c->detection));
193 c->auxiliary =
av_calloc(
s->ar_order + 1,
sizeof(*
c->auxiliary));
194 c->acoefficients =
av_calloc(
s->ar_order + 1,
sizeof(*
c->acoefficients));
195 c->acorrelation =
av_calloc(
s->ar_order + 1,
sizeof(*
c->acorrelation));
197 c->click =
av_calloc(
s->window_size,
sizeof(*
c->click));
198 c->index =
av_calloc(
s->window_size,
sizeof(*
c->index));
199 c->interpolated =
av_calloc(
s->window_size,
sizeof(*
c->interpolated));
200 if (!
c->auxiliary || !
c->acoefficients || !
c->detection || !
c->click ||
201 !
c->index || !
c->interpolated || !
c->acorrelation || !
c->tmp)
209 double *
output,
double scale)
213 for (
i = 0;
i <= order;
i++) {
216 for (j =
i; j <
size; j++)
234 k[0] =
a[0] = -
r[1] /
r[0];
235 alpha =
r[0] * (1. - k[0] * k[0]);
239 for (j = 0; j <
i; j++)
240 epsilon +=
a[j] *
r[
i - j];
245 for (j =
i - 1; j >= 0; j--)
246 k[j] =
a[j] + k[
i] *
a[
i - j - 1];
247 for (j = 0; j <=
i; j++)
279 while (start <=
end) {
280 i = (
end + start) / 2;
296 for (
i = 0;
i < n;
i++) {
297 const int in =
i * n;
301 for (j = 0; j <
i; j++)
302 value -= matrix[j * n + j] * matrix[
in + j] * matrix[
in + j];
309 for (j =
i + 1; j < n; j++) {
310 const int jn = j * n;
314 for (k = 0; k <
i; k++)
315 x -= matrix[k * n + k] * matrix[
in + k] * matrix[jn + k];
316 matrix[jn +
i] = x / matrix[
in +
i];
324 double *vector,
int n,
double *
out)
338 for (
i = 0;
i < n;
i++) {
339 const int in =
i * n;
343 for (j = 0; j <
i; j++)
344 value -= matrix[
in + j] * y[j];
348 for (
i = n - 1;
i >= 0;
i--) {
349 out[
i] = y[
i] / matrix[
i * n +
i];
350 for (j =
i + 1; j < n; j++)
351 out[
i] -= matrix[j * n +
i] *
out[j];
358 double *acoefficients,
int *
index,
int nb_errors,
359 double *auxiliary,
double *interpolated)
361 double *vector, *matrix;
364 av_fast_malloc(&
c->matrix, &
c->matrix_size, nb_errors * nb_errors *
sizeof(*
c->matrix));
376 for (
i = 0;
i < nb_errors;
i++) {
377 const int im =
i * nb_errors;
379 for (j =
i; j < nb_errors; j++) {
381 matrix[j * nb_errors +
i] = matrix[
im + j] = auxiliary[
abs(
index[j] -
index[
i])];
383 matrix[j * nb_errors +
i] = matrix[
im + j] = 0;
388 for (
i = 0;
i < nb_errors;
i++) {
403 double *unused1,
double *unused2,
405 const double *
src,
double *dst)
408 double max_amplitude = 0;
412 av_fast_malloc(&
c->histogram, &
c->histogram_size,
s->nb_hbins *
sizeof(*
c->histogram));
415 histogram =
c->histogram;
416 memset(histogram, 0,
sizeof(*histogram) *
s->nb_hbins);
418 for (
i = 0;
i <
s->window_size;
i++) {
426 for (
i =
s->nb_hbins - 1;
i > 1;
i--) {
429 max_amplitude =
i / (double)
s->nb_hbins;
435 if (max_amplitude > 0.) {
436 for (
i = 0;
i <
s->window_size;
i++) {
441 memset(
clip, 0,
s->ar_order *
sizeof(*
clip));
442 memset(
clip + (
s->window_size -
s->ar_order), 0,
s->ar_order *
sizeof(*
clip));
453 double *detection,
double *acoefficients,
455 const double *
src,
double *dst)
458 int i, j, nb_clicks = 0, prev = -1;
460 memset(detection, 0,
s->window_size *
sizeof(*detection));
463 for (j = 0; j <=
s->ar_order; j++) {
464 detection[
i] += acoefficients[j] *
src[
i - j];
468 for (
i = 0;
i <
s->window_size;
i++) {
469 click[
i] = fabs(detection[
i]) > sigmae *
threshold;
473 for (
i = 0;
i <
s->window_size;
i++) {
478 for (j = prev + 1; j <
i; j++)
483 memset(click, 0,
s->ar_order *
sizeof(*click));
484 memset(click + (
s->window_size -
s->ar_order), 0,
s->ar_order *
sizeof(*click));
502 const double *
src = (
const double *)
s->in->extended_data[ch];
503 double *
is = (
double *)
s->is->extended_data[ch];
504 double *dst = (
double *)
s->out->extended_data[ch];
505 double *ptr = (
double *)
out->extended_data[ch];
506 double *buf = (
double *)
s->buffer->extended_data[ch];
507 const double *
w =
s->window_func_lut;
515 double *interpolated =
c->interpolated;
519 nb_errors =
s->detector(
s,
c, sigmae,
c->detection,
c->acoefficients,
522 double *enabled = (
double *)
s->enabled->extended_data[0];
525 nb_errors,
c->auxiliary, interpolated);
531 for (j = 0; j < nb_errors; j++) {
532 if (enabled[
index[j]]) {
533 dst[
index[j]] = interpolated[j];
539 memcpy(dst,
src,
s->window_size *
sizeof(*dst));
542 if (
s->method == 0) {
543 for (j = 0; j <
s->window_size; j++)
544 buf[j] += dst[j] *
w[j];
546 const int skip =
s->overlap_skip;
548 for (j = 0; j <
s->hop_size; j++)
549 buf[j] = dst[skip + j];
551 for (j = 0; j <
s->hop_size; j++)
554 memmove(buf, buf +
s->hop_size, (
s->window_size * 2 -
s->hop_size) *
sizeof(*buf));
555 memmove(
is,
is +
s->hop_size, (
s->window_size -
s->hop_size) *
sizeof(*
is));
556 memset(buf +
s->window_size * 2 -
s->hop_size, 0,
s->hop_size *
sizeof(*buf));
557 memset(
is +
s->window_size -
s->hop_size, 0,
s->hop_size *
sizeof(*
is));
568 int ret = 0, j, ch, detected_errors = 0;
585 for (ch = 0; ch <
s->in->channels; ch++) {
586 double *
is = (
double *)
s->is->extended_data[ch];
588 for (j = 0; j <
s->hop_size; j++) {
597 if (
s->samples_left > 0)
598 out->nb_samples =
FFMIN(
s->hop_size,
s->samples_left);
603 s->detected_errors += detected_errors;
604 s->nb_samples +=
out->nb_samples *
inlink->channels;
610 if (
s->samples_left > 0) {
611 s->samples_left -=
s->hop_size;
612 if (
s->samples_left <= 0)
637 double *e = (
double *)
s->enabled->extended_data[0];
644 for (
int i = 0;
i <
in->nb_samples;
i++)
645 e[
i] = !
ctx->is_disabled;
671 if (
s->eof &&
s->samples_left <= 0) {
686 s->is_declip = !strcmp(
ctx->filter->name,
"adeclip");
702 s->is_declip ?
"clips" :
"clicks",
s->detected_errors,
703 s->nb_samples, 100. *
s->detected_errors /
s->nb_samples);
715 for (
i = 0;
i <
s->nb_channels;
i++) {
729 c->histogram_size = 0;
762 .priv_class = &adeclick_class,
790 .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.
static av_cold int end(AVCodecContext *avctx)
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.
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)
#define i(width, name, range_min, range_max)
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.