FFmpeg
af_silenceremove.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2001 Heikki Leinonen
3  * Copyright (c) 2001 Chris Bagwell
4  * Copyright (c) 2003 Donnie Smith
5  * Copyright (c) 2014 Paul B Mahol
6  *
7  * This file is part of FFmpeg.
8  *
9  * FFmpeg is free software; you can redistribute it and/or
10  * modify it under the terms of the GNU Lesser General Public
11  * License as published by the Free Software Foundation; either
12  * version 2.1 of the License, or (at your option) any later version.
13  *
14  * FFmpeg is distributed in the hope that it will be useful,
15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17  * Lesser General Public License for more details.
18  *
19  * You should have received a copy of the GNU Lesser General Public
20  * License along with FFmpeg; if not, write to the Free Software
21  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
22  */
23 
24 #include <float.h> /* DBL_MAX */
25 
26 #include "libavutil/opt.h"
27 #include "libavutil/timestamp.h"
28 #include "audio.h"
29 #include "formats.h"
30 #include "avfilter.h"
31 #include "internal.h"
32 
36 };
37 
41 };
42 
49 };
50 
51 typedef struct SilenceRemoveContext {
52  const AVClass *class;
53 
55 
57  int64_t start_duration;
60  int64_t start_silence;
63 
65  int64_t stop_duration;
68  int64_t stop_silence;
70  int stop_mode;
71 
72  double *start_holdoff;
79 
80  double *stop_holdoff;
87 
88  double window_ratio;
89  double *window;
90  double *window_current;
91  double *window_end;
93  double sum;
94 
95  int restart;
96  int64_t next_pts;
97 
98  int detection;
99  void (*update)(struct SilenceRemoveContext *s, double sample);
100  double(*compute)(struct SilenceRemoveContext *s, double sample);
102 
103 #define OFFSET(x) offsetof(SilenceRemoveContext, x)
104 #define AF AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_AUDIO_PARAM
105 
106 static const AVOption silenceremove_options[] = {
107  { "start_periods", NULL, OFFSET(start_periods), AV_OPT_TYPE_INT, {.i64=0}, 0, 9000, AF },
108  { "start_duration", "set start duration of non-silence part", OFFSET(start_duration_opt), AV_OPT_TYPE_DURATION, {.i64=0}, 0, INT32_MAX, AF },
109  { "start_threshold", "set threshold for start silence detection", OFFSET(start_threshold), AV_OPT_TYPE_DOUBLE, {.dbl=0}, 0, DBL_MAX, AF },
110  { "start_silence", "set start duration of silence part to keep", OFFSET(start_silence_opt), AV_OPT_TYPE_DURATION, {.i64=0}, 0, INT32_MAX, AF },
111  { "start_mode", "set which channel will trigger trimming from start", OFFSET(start_mode), AV_OPT_TYPE_INT, {.i64=T_ANY}, T_ANY, T_ALL, AF, "mode" },
112  { "any", 0, 0, AV_OPT_TYPE_CONST, {.i64=T_ANY}, 0, 0, AF, "mode" },
113  { "all", 0, 0, AV_OPT_TYPE_CONST, {.i64=T_ALL}, 0, 0, AF, "mode" },
114  { "stop_periods", NULL, OFFSET(stop_periods), AV_OPT_TYPE_INT, {.i64=0}, -9000, 9000, AF },
115  { "stop_duration", "set stop duration of non-silence part", OFFSET(stop_duration_opt), AV_OPT_TYPE_DURATION, {.i64=0}, 0, INT32_MAX, AF },
116  { "stop_threshold", "set threshold for stop silence detection", OFFSET(stop_threshold), AV_OPT_TYPE_DOUBLE, {.dbl=0}, 0, DBL_MAX, AF },
117  { "stop_silence", "set stop duration of silence part to keep", OFFSET(stop_silence_opt), AV_OPT_TYPE_DURATION, {.i64=0}, 0, INT32_MAX, AF },
118  { "stop_mode", "set which channel will trigger trimming from end", OFFSET(stop_mode), AV_OPT_TYPE_INT, {.i64=T_ANY}, T_ANY, T_ALL, AF, "mode" },
119  { "detection", "set how silence is detected", OFFSET(detection), AV_OPT_TYPE_INT, {.i64=D_RMS}, D_PEAK,D_RMS, AF, "detection" },
120  { "peak", "use absolute values of samples", 0, AV_OPT_TYPE_CONST, {.i64=D_PEAK},0, 0, AF, "detection" },
121  { "rms", "use squared values of samples", 0, AV_OPT_TYPE_CONST, {.i64=D_RMS}, 0, 0, AF, "detection" },
122  { "window", "set duration of window in seconds", OFFSET(window_ratio), AV_OPT_TYPE_DOUBLE, {.dbl=0.02}, 0, 10, AF },
123  { NULL }
124 };
125 
126 AVFILTER_DEFINE_CLASS(silenceremove);
127 
128 static double compute_peak(SilenceRemoveContext *s, double sample)
129 {
130  double new_sum;
131 
132  new_sum = s->sum;
133  new_sum -= *s->window_current;
134  new_sum += fabs(sample);
135 
136  return new_sum / s->window_size;
137 }
138 
140 {
141  s->sum -= *s->window_current;
142  *s->window_current = fabs(sample);
143  s->sum += *s->window_current;
144 
145  s->window_current++;
146  if (s->window_current >= s->window_end)
147  s->window_current = s->window;
148 }
149 
150 static double compute_rms(SilenceRemoveContext *s, double sample)
151 {
152  double new_sum;
153 
154  new_sum = s->sum;
155  new_sum -= *s->window_current;
156  new_sum += sample * sample;
157 
158  return sqrt(new_sum / s->window_size);
159 }
160 
162 {
163  s->sum -= *s->window_current;
164  *s->window_current = sample * sample;
165  s->sum += *s->window_current;
166 
167  s->window_current++;
168  if (s->window_current >= s->window_end)
169  s->window_current = s->window;
170 }
171 
173 {
174  SilenceRemoveContext *s = ctx->priv;
175 
176  if (s->stop_periods < 0) {
177  s->stop_periods = -s->stop_periods;
178  s->restart = 1;
179  }
180 
181  switch (s->detection) {
182  case D_PEAK:
183  s->update = update_peak;
184  s->compute = compute_peak;
185  break;
186  case D_RMS:
187  s->update = update_rms;
188  s->compute = compute_rms;
189  break;
190  }
191 
192  return 0;
193 }
194 
196 {
197  memset(s->window, 0, s->window_size * sizeof(*s->window));
198 
199  s->window_current = s->window;
200  s->window_end = s->window + s->window_size;
201  s->sum = 0;
202 }
203 
205 {
206  AVFilterContext *ctx = inlink->dst;
207  SilenceRemoveContext *s = ctx->priv;
208 
210  s->window_size = FFMAX((inlink->sample_rate * s->window_ratio), 1) * inlink->channels;
211  s->window = av_malloc_array(s->window_size, sizeof(*s->window));
212  if (!s->window)
213  return AVERROR(ENOMEM);
214 
215  clear_window(s);
216 
218  AV_TIME_BASE);
220  AV_TIME_BASE);
222  AV_TIME_BASE);
224  AV_TIME_BASE);
225 
227  sizeof(*s->start_holdoff) *
228  inlink->channels);
229  if (!s->start_holdoff)
230  return AVERROR(ENOMEM);
231 
233  sizeof(*s->start_silence_hold) *
234  inlink->channels);
235  if (!s->start_silence_hold)
236  return AVERROR(ENOMEM);
237 
238  s->start_holdoff_offset = 0;
239  s->start_holdoff_end = 0;
240  s->start_found_periods = 0;
241 
243  sizeof(*s->stop_holdoff) *
244  inlink->channels);
245  if (!s->stop_holdoff)
246  return AVERROR(ENOMEM);
247 
249  sizeof(*s->stop_silence_hold) *
250  inlink->channels);
251  if (!s->stop_silence_hold)
252  return AVERROR(ENOMEM);
253 
254  s->stop_holdoff_offset = 0;
255  s->stop_holdoff_end = 0;
256  s->stop_found_periods = 0;
257 
258  if (s->start_periods)
259  s->mode = SILENCE_TRIM;
260  else
261  s->mode = SILENCE_COPY;
262 
263  return 0;
264 }
265 
267  AVFrame *out, AVFilterLink *outlink,
268  int *nb_samples_written, int *ret, int flush_silence)
269 {
270  AVFrame *silence;
271 
272  if (*nb_samples_written) {
273  out->nb_samples = *nb_samples_written / outlink->channels;
274 
275  out->pts = s->next_pts;
276  s->next_pts += av_rescale_q(out->nb_samples,
277  (AVRational){1, outlink->sample_rate},
278  outlink->time_base);
279 
280  *ret = ff_filter_frame(outlink, out);
281  if (*ret < 0)
282  return;
283  *nb_samples_written = 0;
284  } else {
285  av_frame_free(&out);
286  }
287 
288  if (s->stop_silence_end <= 0 || !flush_silence)
289  return;
290 
291  silence = ff_get_audio_buffer(outlink, s->stop_silence_end / outlink->channels);
292  if (!silence) {
293  *ret = AVERROR(ENOMEM);
294  return;
295  }
296 
298  memcpy(silence->data[0],
300  (s->stop_silence_end - s->stop_silence_offset) * sizeof(double));
301  }
302 
303  if (s->stop_silence_offset > 0) {
304  memcpy(silence->data[0] + (s->stop_silence_end - s->stop_silence_offset) * sizeof(double),
305  &s->stop_silence_hold[0],
306  s->stop_silence_offset * sizeof(double));
307  }
308 
309  s->stop_silence_offset = 0;
310  s->stop_silence_end = 0;
311 
312  silence->pts = s->next_pts;
313  s->next_pts += av_rescale_q(silence->nb_samples,
314  (AVRational){1, outlink->sample_rate},
315  outlink->time_base);
316 
317  *ret = ff_filter_frame(outlink, silence);
318 }
319 
321 {
322  AVFilterContext *ctx = inlink->dst;
323  AVFilterLink *outlink = ctx->outputs[0];
324  SilenceRemoveContext *s = ctx->priv;
325  int i, j, threshold, ret = 0;
326  int nbs, nb_samples_read, nb_samples_written;
327  double *obuf, *ibuf = (double *)in->data[0];
328  AVFrame *out;
329 
330  nb_samples_read = nb_samples_written = 0;
331 
332  if (s->next_pts == AV_NOPTS_VALUE)
333  s->next_pts = in->pts;
334 
335  switch (s->mode) {
336  case SILENCE_TRIM:
337 silence_trim:
338  nbs = in->nb_samples - nb_samples_read / outlink->channels;
339  if (!nbs)
340  break;
341 
342  for (i = 0; i < nbs; i++) {
343  if (s->start_mode == T_ANY) {
344  threshold = 0;
345  for (j = 0; j < outlink->channels; j++) {
346  threshold |= s->compute(s, ibuf[j]) > s->start_threshold;
347  }
348  } else {
349  threshold = 1;
350  for (j = 0; j < outlink->channels; j++) {
351  threshold &= s->compute(s, ibuf[j]) > s->start_threshold;
352  }
353  }
354 
355  if (threshold) {
356  for (j = 0; j < outlink->channels; j++) {
357  s->update(s, *ibuf);
358  s->start_holdoff[s->start_holdoff_end++] = *ibuf++;
359  }
360  nb_samples_read += outlink->channels;
361 
362  if (s->start_holdoff_end >= s->start_duration * outlink->channels) {
363  if (++s->start_found_periods >= s->start_periods) {
365  goto silence_trim_flush;
366  }
367 
368  s->start_holdoff_offset = 0;
369  s->start_holdoff_end = 0;
370  s->start_silence_offset = 0;
371  s->start_silence_end = 0;
372  }
373  } else {
374  s->start_holdoff_end = 0;
375 
376  for (j = 0; j < outlink->channels; j++) {
377  s->update(s, ibuf[j]);
378  if (s->start_silence) {
379  s->start_silence_hold[s->start_silence_offset++] = ibuf[j];
380  s->start_silence_end = FFMIN(s->start_silence_end + 1, outlink->channels * s->start_silence);
381  if (s->start_silence_offset >= outlink->channels * s->start_silence) {
382  s->start_silence_offset = 0;
383  }
384  }
385  }
386 
387  ibuf += outlink->channels;
388  nb_samples_read += outlink->channels;
389  }
390  }
391  break;
392 
393  case SILENCE_TRIM_FLUSH:
394 silence_trim_flush:
396  nbs -= nbs % outlink->channels;
397  if (!nbs)
398  break;
399 
400  out = ff_get_audio_buffer(outlink, nbs / outlink->channels + s->start_silence_end / outlink->channels);
401  if (!out) {
402  av_frame_free(&in);
403  return AVERROR(ENOMEM);
404  }
405 
406  if (s->start_silence_end > 0) {
408  memcpy(out->data[0],
410  (s->start_silence_end - s->start_silence_offset) * sizeof(double));
411  }
412 
413  if (s->start_silence_offset > 0) {
414  memcpy(out->data[0] + (s->start_silence_end - s->start_silence_offset) * sizeof(double),
415  &s->start_silence_hold[0],
416  s->start_silence_offset * sizeof(double));
417  }
418  }
419 
420  memcpy(out->data[0] + s->start_silence_end * sizeof(double),
422  nbs * sizeof(double));
423 
424  out->pts = s->next_pts;
425  s->next_pts += av_rescale_q(out->nb_samples,
426  (AVRational){1, outlink->sample_rate},
427  outlink->time_base);
428 
429  s->start_holdoff_offset += nbs;
430 
431  ret = ff_filter_frame(outlink, out);
432 
434  s->start_holdoff_offset = 0;
435  s->start_holdoff_end = 0;
436  s->start_silence_offset = 0;
437  s->start_silence_end = 0;
438  s->mode = SILENCE_COPY;
439  goto silence_copy;
440  }
441  break;
442 
443  case SILENCE_COPY:
444 silence_copy:
445  nbs = in->nb_samples - nb_samples_read / outlink->channels;
446  if (!nbs)
447  break;
448 
449  out = ff_get_audio_buffer(outlink, nbs);
450  if (!out) {
451  av_frame_free(&in);
452  return AVERROR(ENOMEM);
453  }
454  obuf = (double *)out->data[0];
455 
456  if (s->stop_periods) {
457  for (i = 0; i < nbs; i++) {
458  if (s->stop_mode == T_ANY) {
459  threshold = 0;
460  for (j = 0; j < outlink->channels; j++) {
461  threshold |= s->compute(s, ibuf[j]) > s->stop_threshold;
462  }
463  } else {
464  threshold = 1;
465  for (j = 0; j < outlink->channels; j++) {
466  threshold &= s->compute(s, ibuf[j]) > s->stop_threshold;
467  }
468  }
469 
470  if (threshold && s->stop_holdoff_end && !s->stop_silence) {
472  flush(s, out, outlink, &nb_samples_written, &ret, 0);
473  goto silence_copy_flush;
474  } else if (threshold) {
475  for (j = 0; j < outlink->channels; j++) {
476  s->update(s, *ibuf);
477  *obuf++ = *ibuf++;
478  }
479  nb_samples_read += outlink->channels;
480  nb_samples_written += outlink->channels;
481  } else if (!threshold) {
482  for (j = 0; j < outlink->channels; j++) {
483  s->update(s, *ibuf);
484  if (s->stop_silence) {
485  s->stop_silence_hold[s->stop_silence_offset++] = *ibuf;
486  s->stop_silence_end = FFMIN(s->stop_silence_end + 1, outlink->channels * s->stop_silence);
487  if (s->stop_silence_offset >= outlink->channels * s->stop_silence) {
488  s->stop_silence_offset = 0;
489  }
490  }
491 
492  s->stop_holdoff[s->stop_holdoff_end++] = *ibuf++;
493  }
494  nb_samples_read += outlink->channels;
495 
496  if (s->stop_holdoff_end >= s->stop_duration * outlink->channels) {
497  if (++s->stop_found_periods >= s->stop_periods) {
498  s->stop_holdoff_offset = 0;
499  s->stop_holdoff_end = 0;
500 
501  if (!s->restart) {
502  s->mode = SILENCE_STOP;
503  flush(s, out, outlink, &nb_samples_written, &ret, 1);
504  goto silence_stop;
505  } else {
506  s->stop_found_periods = 0;
507  s->start_found_periods = 0;
508  s->start_holdoff_offset = 0;
509  s->start_holdoff_end = 0;
510  s->start_silence_offset = 0;
511  s->start_silence_end = 0;
512  clear_window(s);
513  s->mode = SILENCE_TRIM;
514  flush(s, out, outlink, &nb_samples_written, &ret, 1);
515  goto silence_trim;
516  }
517  }
519  flush(s, out, outlink, &nb_samples_written, &ret, 0);
520  goto silence_copy_flush;
521  }
522  }
523  }
524  flush(s, out, outlink, &nb_samples_written, &ret, 0);
525  } else {
526  memcpy(obuf, ibuf, sizeof(double) * nbs * outlink->channels);
527 
528  out->pts = s->next_pts;
529  s->next_pts += av_rescale_q(out->nb_samples,
530  (AVRational){1, outlink->sample_rate},
531  outlink->time_base);
532 
533  ret = ff_filter_frame(outlink, out);
534  }
535  break;
536 
537  case SILENCE_COPY_FLUSH:
538 silence_copy_flush:
540  nbs -= nbs % outlink->channels;
541  if (!nbs)
542  break;
543 
544  out = ff_get_audio_buffer(outlink, nbs / outlink->channels);
545  if (!out) {
546  av_frame_free(&in);
547  return AVERROR(ENOMEM);
548  }
549 
550  memcpy(out->data[0], &s->stop_holdoff[s->stop_holdoff_offset],
551  nbs * sizeof(double));
552  s->stop_holdoff_offset += nbs;
553 
554  out->pts = s->next_pts;
555  s->next_pts += av_rescale_q(out->nb_samples,
556  (AVRational){1, outlink->sample_rate},
557  outlink->time_base);
558 
559  ret = ff_filter_frame(outlink, out);
560 
561  if (s->stop_holdoff_offset == s->stop_holdoff_end) {
562  s->stop_holdoff_offset = 0;
563  s->stop_holdoff_end = 0;
564  s->stop_silence_offset = 0;
565  s->stop_silence_end = 0;
566  s->mode = SILENCE_COPY;
567  goto silence_copy;
568  }
569  break;
570  case SILENCE_STOP:
571 silence_stop:
572  break;
573  }
574 
575  av_frame_free(&in);
576 
577  return ret;
578 }
579 
580 static int request_frame(AVFilterLink *outlink)
581 {
582  AVFilterContext *ctx = outlink->src;
583  SilenceRemoveContext *s = ctx->priv;
584  int ret;
585 
586  ret = ff_request_frame(ctx->inputs[0]);
587  if (ret == AVERROR_EOF && (s->mode == SILENCE_COPY_FLUSH ||
588  s->mode == SILENCE_COPY)) {
589  int nbs = s->stop_holdoff_end - s->stop_holdoff_offset;
590  if (nbs) {
591  AVFrame *frame;
592 
593  frame = ff_get_audio_buffer(outlink, nbs / outlink->channels);
594  if (!frame)
595  return AVERROR(ENOMEM);
596 
597  memcpy(frame->data[0], &s->stop_holdoff[s->stop_holdoff_offset],
598  nbs * sizeof(double));
599 
600  frame->pts = s->next_pts;
601  s->next_pts += av_rescale_q(frame->nb_samples,
602  (AVRational){1, outlink->sample_rate},
603  outlink->time_base);
604 
605  ret = ff_filter_frame(outlink, frame);
606  }
607  s->mode = SILENCE_STOP;
608  }
609  return ret;
610 }
611 
613 {
616  static const enum AVSampleFormat sample_fmts[] = {
618  };
619  int ret;
620 
621  layouts = ff_all_channel_counts();
622  if (!layouts)
623  return AVERROR(ENOMEM);
624  ret = ff_set_common_channel_layouts(ctx, layouts);
625  if (ret < 0)
626  return ret;
627 
628  formats = ff_make_format_list(sample_fmts);
629  if (!formats)
630  return AVERROR(ENOMEM);
631  ret = ff_set_common_formats(ctx, formats);
632  if (ret < 0)
633  return ret;
634 
635  formats = ff_all_samplerates();
636  if (!formats)
637  return AVERROR(ENOMEM);
638  return ff_set_common_samplerates(ctx, formats);
639 }
640 
642 {
643  SilenceRemoveContext *s = ctx->priv;
644 
645  av_freep(&s->start_holdoff);
647  av_freep(&s->stop_holdoff);
649  av_freep(&s->window);
650 }
651 
653  {
654  .name = "default",
655  .type = AVMEDIA_TYPE_AUDIO,
656  .config_props = config_input,
657  .filter_frame = filter_frame,
658  },
659  { NULL }
660 };
661 
663  {
664  .name = "default",
665  .type = AVMEDIA_TYPE_AUDIO,
666  .request_frame = request_frame,
667  },
668  { NULL }
669 };
670 
672  .name = "silenceremove",
673  .description = NULL_IF_CONFIG_SMALL("Remove silence."),
674  .priv_size = sizeof(SilenceRemoveContext),
675  .priv_class = &silenceremove_class,
676  .init = init,
677  .uninit = uninit,
679  .inputs = silenceremove_inputs,
680  .outputs = silenceremove_outputs,
681 };
static int filter_frame(AVFilterLink *inlink, AVFrame *in)
#define NULL
Definition: coverity.c:32
int ff_set_common_channel_layouts(AVFilterContext *ctx, AVFilterChannelLayouts *layouts)
A helper for query_formats() which sets all links to the same list of channel layouts/sample rates...
Definition: formats.c:549
void(* update)(struct SilenceRemoveContext *s, double sample)
This structure describes decoded (raw) audio or video data.
Definition: frame.h:295
SilenceDetect
AVOption.
Definition: opt.h:246
static void flush(SilenceRemoveContext *s, AVFrame *out, AVFilterLink *outlink, int *nb_samples_written, int *ret, int flush_silence)
static const AVFilterPad silenceremove_outputs[]
Main libavfilter public API header.
AVFilter ff_af_silenceremove
#define sample
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:283
const char * name
Pad name.
Definition: internal.h:60
AVFilterLink ** inputs
array of pointers to input links
Definition: avfilter.h:346
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1080
#define av_cold
Definition: attributes.h:82
AVOptions.
timestamp utils, mostly useful for debugging/logging purposes
SilenceMode
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:388
static av_cold int init(AVFilterContext *ctx)
#define AVERROR_EOF
End of file.
Definition: error.h:55
A filter pad used for either input or output.
Definition: internal.h:54
int64_t av_rescale_q(int64_t a, AVRational bq, AVRational cq)
Rescale a 64-bit integer by 2 rational numbers.
Definition: mathematics.c:142
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:259
int ff_set_common_formats(AVFilterContext *ctx, AVFilterFormats *formats)
A helper for query_formats() which sets all links to the same list of formats.
Definition: formats.c:568
AVFrame * ff_get_audio_buffer(AVFilterLink *link, int nb_samples)
Request an audio samples buffer with a specific set of permissions.
Definition: audio.c:86
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:202
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:186
void * priv
private data for use by the filter
Definition: avfilter.h:353
#define FFMAX(a, b)
Definition: common.h:94
ThresholdMode
double(* compute)(struct SilenceRemoveContext *s, double sample)
static int query_formats(AVFilterContext *ctx)
static int config_input(AVFilterLink *inlink)
int64_t av_rescale(int64_t a, int64_t b, int64_t c)
Rescale a 64-bit integer with rounding to nearest.
Definition: mathematics.c:129
#define AV_TIME_BASE
Internal time base represented as integer.
Definition: avutil.h:254
#define FFMIN(a, b)
Definition: common.h:96
static const AVFilterPad silenceremove_inputs[]
typedef void(APIENTRY *FF_PFNGLACTIVETEXTUREPROC)(GLenum texture)
AVFormatContext * ctx
Definition: movenc.c:48
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several the filter must be ready for frames arriving randomly on any input any filter with several inputs will most likely require some kind of queuing mechanism It is perfectly acceptable to have a limited queue and to drop frames when the inputs are too unbalanced request_frame For filters that do not use the this method is called when a frame is wanted on an output For a it should directly call filter_frame on the corresponding output For a if there are queued frames already one of these frames should be pushed If the filter should request a frame on one of its repeatedly until at least one frame has been pushed Return or at least make progress towards producing a frame
#define s(width, name)
Definition: cbs_vp9.c:257
static int request_frame(AVFilterLink *outlink)
AVFILTER_DEFINE_CLASS(silenceremove)
enum SilenceMode mode
static const AVFilterPad outputs[]
Definition: af_acontrast.c:203
A list of supported channel layouts.
Definition: formats.h:85
if(ret)
static double compute_peak(SilenceRemoveContext *s, double sample)
AVSampleFormat
Audio sample formats.
Definition: samplefmt.h:58
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several inputs
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 av_cold void uninit(AVFilterContext *ctx)
Describe the class of an AVClass context structure.
Definition: log.h:67
Filter definition.
Definition: avfilter.h:144
static void update_rms(SilenceRemoveContext *s, double sample)
Rational number (pair of numerator and denominator).
Definition: rational.h:58
static double compute_rms(SilenceRemoveContext *s, double sample)
static void clear_window(SilenceRemoveContext *s)
#define AF
const char * name
Filter name.
Definition: avfilter.h:148
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:350
enum MovChannelLayoutTag * layouts
Definition: mov_chan.c:434
AVFilterFormats * ff_all_samplerates(void)
Definition: formats.c:395
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:309
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
static void update_peak(SilenceRemoveContext *s, double sample)
static const AVOption silenceremove_options[]
A list of supported formats for one end of a filter link.
Definition: formats.h:64
An instance of a filter.
Definition: avfilter.h:338
static enum AVSampleFormat sample_fmts[]
Definition: adpcmenc.c:701
FILE * out
Definition: movenc.c:54
#define av_freep(p)
#define av_malloc_array(a, b)
int ff_request_frame(AVFilterLink *link)
Request an input frame from the filter at the other end of the link.
Definition: avfilter.c:407
formats
Definition: signature.h:48
internal API functions
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
AVFilterChannelLayouts * ff_all_channel_counts(void)
Construct an AVFilterChannelLayouts coding for any channel layout, with known or unknown disposition...
Definition: formats.c:410
#define OFFSET(x)
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:361
int ff_set_common_samplerates(AVFilterContext *ctx, AVFilterFormats *samplerates)
Definition: formats.c:556
#define AV_NOPTS_VALUE
Undefined timestamp value.
Definition: avutil.h:248