FFmpeg
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
af_atempo.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2012 Pavel Koshevoy <pkoshevoy at gmail dot com>
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or
7  * modify it under the terms of the GNU Lesser General Public
8  * License as published by the Free Software Foundation; either
9  * version 2.1 of the License, or (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14  * Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with FFmpeg; if not, write to the Free Software
18  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19  */
20 
21 /**
22  * @file
23  * tempo scaling audio filter -- an implementation of WSOLA algorithm
24  *
25  * Based on MIT licensed yaeAudioTempoFilter.h and yaeAudioFragment.h
26  * from Apprentice Video player by Pavel Koshevoy.
27  * https://sourceforge.net/projects/apprenticevideo/
28  *
29  * An explanation of SOLA algorithm is available at
30  * http://www.surina.net/article/time-and-pitch-scaling.html
31  *
32  * WSOLA is very similar to SOLA, only one major difference exists between
33  * these algorithms. SOLA shifts audio fragments along the output stream,
34  * where as WSOLA shifts audio fragments along the input stream.
35  *
36  * The advantage of WSOLA algorithm is that the overlap region size is
37  * always the same, therefore the blending function is constant and
38  * can be precomputed.
39  */
40 
41 #include <float.h>
42 #include "libavcodec/avfft.h"
43 #include "libavutil/avassert.h"
44 #include "libavutil/avstring.h"
46 #include "libavutil/eval.h"
47 #include "libavutil/opt.h"
48 #include "libavutil/samplefmt.h"
49 #include "avfilter.h"
50 #include "audio.h"
51 #include "internal.h"
52 
53 /**
54  * A fragment of audio waveform
55  */
56 typedef struct AudioFragment {
57  // index of the first sample of this fragment in the overall waveform;
58  // 0: input sample position
59  // 1: output sample position
60  int64_t position[2];
61 
62  // original packed multi-channel samples:
64 
65  // number of samples in this fragment:
66  int nsamples;
67 
68  // rDFT transform of the down-mixed mono fragment, used for
69  // fast waveform alignment via correlation in frequency domain:
72 
73 /**
74  * Filter state machine states
75  */
76 typedef enum {
82 } FilterState;
83 
84 /**
85  * Filter state machine
86  */
87 typedef struct ATempoContext {
88  const AVClass *class;
89 
90  // ring-buffer of input samples, necessary because some times
91  // input fragment position may be adjusted backwards:
93 
94  // ring-buffer maximum capacity, expressed in sample rate time base:
95  int ring;
96 
97  // ring-buffer house keeping:
98  int size;
99  int head;
100  int tail;
101 
102  // 0: input sample position corresponding to the ring buffer tail
103  // 1: output sample position
104  int64_t position[2];
105 
106  // sample format:
108 
109  // number of channels:
110  int channels;
111 
112  // row of bytes to skip from one sample to next, across multple channels;
113  // stride = (number-of-channels * bits-per-sample-per-channel) / 8
114  int stride;
115 
116  // fragment window size, power-of-two integer:
117  int window;
118 
119  // Hann window coefficients, for feathering
120  // (blending) the overlapping fragment region:
121  float *hann;
122 
123  // tempo scaling factor:
124  double tempo;
125 
126  // a snapshot of previous fragment input and output position values
127  // captured when the tempo scale factor was set most recently:
128  int64_t origin[2];
129 
130  // current/previous fragment ring-buffer:
132 
133  // current fragment index:
134  uint64_t nfrag;
135 
136  // current state:
138 
139  // for fast correlation calculation in frequency domain:
143 
144  // for managing AVFilterPad.request_frame and AVFilterPad.filter_frame
148  uint64_t nsamples_in;
149  uint64_t nsamples_out;
150 } ATempoContext;
151 
152 #define YAE_ATEMPO_MIN 0.5
153 #define YAE_ATEMPO_MAX 100.0
154 
155 #define OFFSET(x) offsetof(ATempoContext, x)
156 
157 static const AVOption atempo_options[] = {
158  { "tempo", "set tempo scale factor",
159  OFFSET(tempo), AV_OPT_TYPE_DOUBLE, { .dbl = 1.0 },
163  { NULL }
164 };
165 
166 AVFILTER_DEFINE_CLASS(atempo);
167 
169 {
170  return &atempo->frag[atempo->nfrag % 2];
171 }
172 
174 {
175  return &atempo->frag[(atempo->nfrag + 1) % 2];
176 }
177 
178 /**
179  * Reset filter to initial state, do not deallocate existing local buffers.
180  */
181 static void yae_clear(ATempoContext *atempo)
182 {
183  atempo->size = 0;
184  atempo->head = 0;
185  atempo->tail = 0;
186 
187  atempo->nfrag = 0;
188  atempo->state = YAE_LOAD_FRAGMENT;
189 
190  atempo->position[0] = 0;
191  atempo->position[1] = 0;
192 
193  atempo->origin[0] = 0;
194  atempo->origin[1] = 0;
195 
196  atempo->frag[0].position[0] = 0;
197  atempo->frag[0].position[1] = 0;
198  atempo->frag[0].nsamples = 0;
199 
200  atempo->frag[1].position[0] = 0;
201  atempo->frag[1].position[1] = 0;
202  atempo->frag[1].nsamples = 0;
203 
204  // shift left position of 1st fragment by half a window
205  // so that no re-normalization would be required for
206  // the left half of the 1st fragment:
207  atempo->frag[0].position[0] = -(int64_t)(atempo->window / 2);
208  atempo->frag[0].position[1] = -(int64_t)(atempo->window / 2);
209 
210  av_frame_free(&atempo->dst_buffer);
211  atempo->dst = NULL;
212  atempo->dst_end = NULL;
213 
214  atempo->nsamples_in = 0;
215  atempo->nsamples_out = 0;
216 }
217 
218 /**
219  * Reset filter to initial state and deallocate all buffers.
220  */
221 static void yae_release_buffers(ATempoContext *atempo)
222 {
223  yae_clear(atempo);
224 
225  av_freep(&atempo->frag[0].data);
226  av_freep(&atempo->frag[1].data);
227  av_freep(&atempo->frag[0].xdat);
228  av_freep(&atempo->frag[1].xdat);
229 
230  av_freep(&atempo->buffer);
231  av_freep(&atempo->hann);
232  av_freep(&atempo->correlation);
233 
234  av_rdft_end(atempo->real_to_complex);
235  atempo->real_to_complex = NULL;
236 
237  av_rdft_end(atempo->complex_to_real);
238  atempo->complex_to_real = NULL;
239 }
240 
241 /* av_realloc is not aligned enough; fortunately, the data does not need to
242  * be preserved */
243 #define RE_MALLOC_OR_FAIL(field, field_size) \
244  do { \
245  av_freep(&field); \
246  field = av_malloc(field_size); \
247  if (!field) { \
248  yae_release_buffers(atempo); \
249  return AVERROR(ENOMEM); \
250  } \
251  } while (0)
252 
253 /**
254  * Prepare filter for processing audio data of given format,
255  * sample rate and number of channels.
256  */
257 static int yae_reset(ATempoContext *atempo,
258  enum AVSampleFormat format,
259  int sample_rate,
260  int channels)
261 {
262  const int sample_size = av_get_bytes_per_sample(format);
263  uint32_t nlevels = 0;
264  uint32_t pot;
265  int i;
266 
267  atempo->format = format;
268  atempo->channels = channels;
269  atempo->stride = sample_size * channels;
270 
271  // pick a segment window size:
272  atempo->window = sample_rate / 24;
273 
274  // adjust window size to be a power-of-two integer:
275  nlevels = av_log2(atempo->window);
276  pot = 1 << nlevels;
277  av_assert0(pot <= atempo->window);
278 
279  if (pot < atempo->window) {
280  atempo->window = pot * 2;
281  nlevels++;
282  }
283 
284  // initialize audio fragment buffers:
285  RE_MALLOC_OR_FAIL(atempo->frag[0].data, atempo->window * atempo->stride);
286  RE_MALLOC_OR_FAIL(atempo->frag[1].data, atempo->window * atempo->stride);
287  RE_MALLOC_OR_FAIL(atempo->frag[0].xdat, atempo->window * sizeof(FFTComplex));
288  RE_MALLOC_OR_FAIL(atempo->frag[1].xdat, atempo->window * sizeof(FFTComplex));
289 
290  // initialize rDFT contexts:
291  av_rdft_end(atempo->real_to_complex);
292  atempo->real_to_complex = NULL;
293 
294  av_rdft_end(atempo->complex_to_real);
295  atempo->complex_to_real = NULL;
296 
297  atempo->real_to_complex = av_rdft_init(nlevels + 1, DFT_R2C);
298  if (!atempo->real_to_complex) {
299  yae_release_buffers(atempo);
300  return AVERROR(ENOMEM);
301  }
302 
303  atempo->complex_to_real = av_rdft_init(nlevels + 1, IDFT_C2R);
304  if (!atempo->complex_to_real) {
305  yae_release_buffers(atempo);
306  return AVERROR(ENOMEM);
307  }
308 
309  RE_MALLOC_OR_FAIL(atempo->correlation, atempo->window * sizeof(FFTComplex));
310 
311  atempo->ring = atempo->window * 3;
312  RE_MALLOC_OR_FAIL(atempo->buffer, atempo->ring * atempo->stride);
313 
314  // initialize the Hann window function:
315  RE_MALLOC_OR_FAIL(atempo->hann, atempo->window * sizeof(float));
316 
317  for (i = 0; i < atempo->window; i++) {
318  double t = (double)i / (double)(atempo->window - 1);
319  double h = 0.5 * (1.0 - cos(2.0 * M_PI * t));
320  atempo->hann[i] = (float)h;
321  }
322 
323  yae_clear(atempo);
324  return 0;
325 }
326 
327 static int yae_set_tempo(AVFilterContext *ctx, const char *arg_tempo)
328 {
329  const AudioFragment *prev;
330  ATempoContext *atempo = ctx->priv;
331  char *tail = NULL;
332  double tempo = av_strtod(arg_tempo, &tail);
333 
334  if (tail && *tail) {
335  av_log(ctx, AV_LOG_ERROR, "Invalid tempo value '%s'\n", arg_tempo);
336  return AVERROR(EINVAL);
337  }
338 
339  if (tempo < YAE_ATEMPO_MIN || tempo > YAE_ATEMPO_MAX) {
340  av_log(ctx, AV_LOG_ERROR, "Tempo value %f exceeds [%f, %f] range\n",
341  tempo, YAE_ATEMPO_MIN, YAE_ATEMPO_MAX);
342  return AVERROR(EINVAL);
343  }
344 
345  prev = yae_prev_frag(atempo);
346  atempo->origin[0] = prev->position[0] + atempo->window / 2;
347  atempo->origin[1] = prev->position[1] + atempo->window / 2;
348  atempo->tempo = tempo;
349  return 0;
350 }
351 
352 /**
353  * A helper macro for initializing complex data buffer with scalar data
354  * of a given type.
355  */
356 #define yae_init_xdat(scalar_type, scalar_max) \
357  do { \
358  const uint8_t *src_end = src + \
359  frag->nsamples * atempo->channels * sizeof(scalar_type); \
360  \
361  FFTSample *xdat = frag->xdat; \
362  scalar_type tmp; \
363  \
364  if (atempo->channels == 1) { \
365  for (; src < src_end; xdat++) { \
366  tmp = *(const scalar_type *)src; \
367  src += sizeof(scalar_type); \
368  \
369  *xdat = (FFTSample)tmp; \
370  } \
371  } else { \
372  FFTSample s, max, ti, si; \
373  int i; \
374  \
375  for (; src < src_end; xdat++) { \
376  tmp = *(const scalar_type *)src; \
377  src += sizeof(scalar_type); \
378  \
379  max = (FFTSample)tmp; \
380  s = FFMIN((FFTSample)scalar_max, \
381  (FFTSample)fabsf(max)); \
382  \
383  for (i = 1; i < atempo->channels; i++) { \
384  tmp = *(const scalar_type *)src; \
385  src += sizeof(scalar_type); \
386  \
387  ti = (FFTSample)tmp; \
388  si = FFMIN((FFTSample)scalar_max, \
389  (FFTSample)fabsf(ti)); \
390  \
391  if (s < si) { \
392  s = si; \
393  max = ti; \
394  } \
395  } \
396  \
397  *xdat = max; \
398  } \
399  } \
400  } while (0)
401 
402 /**
403  * Initialize complex data buffer of a given audio fragment
404  * with down-mixed mono data of appropriate scalar type.
405  */
406 static void yae_downmix(ATempoContext *atempo, AudioFragment *frag)
407 {
408  // shortcuts:
409  const uint8_t *src = frag->data;
410 
411  // init complex data buffer used for FFT and Correlation:
412  memset(frag->xdat, 0, sizeof(FFTComplex) * atempo->window);
413 
414  if (atempo->format == AV_SAMPLE_FMT_U8) {
415  yae_init_xdat(uint8_t, 127);
416  } else if (atempo->format == AV_SAMPLE_FMT_S16) {
417  yae_init_xdat(int16_t, 32767);
418  } else if (atempo->format == AV_SAMPLE_FMT_S32) {
419  yae_init_xdat(int, 2147483647);
420  } else if (atempo->format == AV_SAMPLE_FMT_FLT) {
421  yae_init_xdat(float, 1);
422  } else if (atempo->format == AV_SAMPLE_FMT_DBL) {
423  yae_init_xdat(double, 1);
424  }
425 }
426 
427 /**
428  * Populate the internal data buffer on as-needed basis.
429  *
430  * @return
431  * 0 if requested data was already available or was successfully loaded,
432  * AVERROR(EAGAIN) if more input data is required.
433  */
434 static int yae_load_data(ATempoContext *atempo,
435  const uint8_t **src_ref,
436  const uint8_t *src_end,
437  int64_t stop_here)
438 {
439  // shortcut:
440  const uint8_t *src = *src_ref;
441  const int read_size = stop_here - atempo->position[0];
442 
443  if (stop_here <= atempo->position[0]) {
444  return 0;
445  }
446 
447  // samples are not expected to be skipped, unless tempo is greater than 2:
448  av_assert0(read_size <= atempo->ring || atempo->tempo > 2.0);
449 
450  while (atempo->position[0] < stop_here && src < src_end) {
451  int src_samples = (src_end - src) / atempo->stride;
452 
453  // load data piece-wise, in order to avoid complicating the logic:
454  int nsamples = FFMIN(read_size, src_samples);
455  int na;
456  int nb;
457 
458  nsamples = FFMIN(nsamples, atempo->ring);
459  na = FFMIN(nsamples, atempo->ring - atempo->tail);
460  nb = FFMIN(nsamples - na, atempo->ring);
461 
462  if (na) {
463  uint8_t *a = atempo->buffer + atempo->tail * atempo->stride;
464  memcpy(a, src, na * atempo->stride);
465 
466  src += na * atempo->stride;
467  atempo->position[0] += na;
468 
469  atempo->size = FFMIN(atempo->size + na, atempo->ring);
470  atempo->tail = (atempo->tail + na) % atempo->ring;
471  atempo->head =
472  atempo->size < atempo->ring ?
473  atempo->tail - atempo->size :
474  atempo->tail;
475  }
476 
477  if (nb) {
478  uint8_t *b = atempo->buffer;
479  memcpy(b, src, nb * atempo->stride);
480 
481  src += nb * atempo->stride;
482  atempo->position[0] += nb;
483 
484  atempo->size = FFMIN(atempo->size + nb, atempo->ring);
485  atempo->tail = (atempo->tail + nb) % atempo->ring;
486  atempo->head =
487  atempo->size < atempo->ring ?
488  atempo->tail - atempo->size :
489  atempo->tail;
490  }
491  }
492 
493  // pass back the updated source buffer pointer:
494  *src_ref = src;
495 
496  // sanity check:
497  av_assert0(atempo->position[0] <= stop_here);
498 
499  return atempo->position[0] == stop_here ? 0 : AVERROR(EAGAIN);
500 }
501 
502 /**
503  * Populate current audio fragment data buffer.
504  *
505  * @return
506  * 0 when the fragment is ready,
507  * AVERROR(EAGAIN) if more input data is required.
508  */
509 static int yae_load_frag(ATempoContext *atempo,
510  const uint8_t **src_ref,
511  const uint8_t *src_end)
512 {
513  // shortcuts:
514  AudioFragment *frag = yae_curr_frag(atempo);
515  uint8_t *dst;
516  int64_t missing, start, zeros;
517  uint32_t nsamples;
518  const uint8_t *a, *b;
519  int i0, i1, n0, n1, na, nb;
520 
521  int64_t stop_here = frag->position[0] + atempo->window;
522  if (src_ref && yae_load_data(atempo, src_ref, src_end, stop_here) != 0) {
523  return AVERROR(EAGAIN);
524  }
525 
526  // calculate the number of samples we don't have:
527  missing =
528  stop_here > atempo->position[0] ?
529  stop_here - atempo->position[0] : 0;
530 
531  nsamples =
532  missing < (int64_t)atempo->window ?
533  (uint32_t)(atempo->window - missing) : 0;
534 
535  // setup the output buffer:
536  frag->nsamples = nsamples;
537  dst = frag->data;
538 
539  start = atempo->position[0] - atempo->size;
540  zeros = 0;
541 
542  if (frag->position[0] < start) {
543  // what we don't have we substitute with zeros:
544  zeros = FFMIN(start - frag->position[0], (int64_t)nsamples);
545  av_assert0(zeros != nsamples);
546 
547  memset(dst, 0, zeros * atempo->stride);
548  dst += zeros * atempo->stride;
549  }
550 
551  if (zeros == nsamples) {
552  return 0;
553  }
554 
555  // get the remaining data from the ring buffer:
556  na = (atempo->head < atempo->tail ?
557  atempo->tail - atempo->head :
558  atempo->ring - atempo->head);
559 
560  nb = atempo->head < atempo->tail ? 0 : atempo->tail;
561 
562  // sanity check:
563  av_assert0(nsamples <= zeros + na + nb);
564 
565  a = atempo->buffer + atempo->head * atempo->stride;
566  b = atempo->buffer;
567 
568  i0 = frag->position[0] + zeros - start;
569  i1 = i0 < na ? 0 : i0 - na;
570 
571  n0 = i0 < na ? FFMIN(na - i0, (int)(nsamples - zeros)) : 0;
572  n1 = nsamples - zeros - n0;
573 
574  if (n0) {
575  memcpy(dst, a + i0 * atempo->stride, n0 * atempo->stride);
576  dst += n0 * atempo->stride;
577  }
578 
579  if (n1) {
580  memcpy(dst, b + i1 * atempo->stride, n1 * atempo->stride);
581  }
582 
583  return 0;
584 }
585 
586 /**
587  * Prepare for loading next audio fragment.
588  */
590 {
591  const double fragment_step = atempo->tempo * (double)(atempo->window / 2);
592 
593  const AudioFragment *prev;
594  AudioFragment *frag;
595 
596  atempo->nfrag++;
597  prev = yae_prev_frag(atempo);
598  frag = yae_curr_frag(atempo);
599 
600  frag->position[0] = prev->position[0] + (int64_t)fragment_step;
601  frag->position[1] = prev->position[1] + atempo->window / 2;
602  frag->nsamples = 0;
603 }
604 
605 /**
606  * Calculate cross-correlation via rDFT.
607  *
608  * Multiply two vectors of complex numbers (result of real_to_complex rDFT)
609  * and transform back via complex_to_real rDFT.
610  */
611 static void yae_xcorr_via_rdft(FFTSample *xcorr,
612  RDFTContext *complex_to_real,
613  const FFTComplex *xa,
614  const FFTComplex *xb,
615  const int window)
616 {
617  FFTComplex *xc = (FFTComplex *)xcorr;
618  int i;
619 
620  // NOTE: first element requires special care -- Given Y = rDFT(X),
621  // Im(Y[0]) and Im(Y[N/2]) are always zero, therefore av_rdft_calc
622  // stores Re(Y[N/2]) in place of Im(Y[0]).
623 
624  xc->re = xa->re * xb->re;
625  xc->im = xa->im * xb->im;
626  xa++;
627  xb++;
628  xc++;
629 
630  for (i = 1; i < window; i++, xa++, xb++, xc++) {
631  xc->re = (xa->re * xb->re + xa->im * xb->im);
632  xc->im = (xa->im * xb->re - xa->re * xb->im);
633  }
634 
635  // apply inverse rDFT:
636  av_rdft_calc(complex_to_real, xcorr);
637 }
638 
639 /**
640  * Calculate alignment offset for given fragment
641  * relative to the previous fragment.
642  *
643  * @return alignment offset of current fragment relative to previous.
644  */
645 static int yae_align(AudioFragment *frag,
646  const AudioFragment *prev,
647  const int window,
648  const int delta_max,
649  const int drift,
651  RDFTContext *complex_to_real)
652 {
653  int best_offset = -drift;
654  FFTSample best_metric = -FLT_MAX;
655  FFTSample *xcorr;
656 
657  int i0;
658  int i1;
659  int i;
660 
661  yae_xcorr_via_rdft(correlation,
662  complex_to_real,
663  (const FFTComplex *)prev->xdat,
664  (const FFTComplex *)frag->xdat,
665  window);
666 
667  // identify search window boundaries:
668  i0 = FFMAX(window / 2 - delta_max - drift, 0);
669  i0 = FFMIN(i0, window);
670 
671  i1 = FFMIN(window / 2 + delta_max - drift, window - window / 16);
672  i1 = FFMAX(i1, 0);
673 
674  // identify cross-correlation peaks within search window:
675  xcorr = correlation + i0;
676 
677  for (i = i0; i < i1; i++, xcorr++) {
678  FFTSample metric = *xcorr;
679 
680  // normalize:
681  FFTSample drifti = (FFTSample)(drift + i);
682  metric *= drifti * (FFTSample)(i - i0) * (FFTSample)(i1 - i);
683 
684  if (metric > best_metric) {
685  best_metric = metric;
686  best_offset = i - window / 2;
687  }
688  }
689 
690  return best_offset;
691 }
692 
693 /**
694  * Adjust current fragment position for better alignment
695  * with previous fragment.
696  *
697  * @return alignment correction.
698  */
700 {
701  const AudioFragment *prev = yae_prev_frag(atempo);
702  AudioFragment *frag = yae_curr_frag(atempo);
703 
704  const double prev_output_position =
705  (double)(prev->position[1] - atempo->origin[1] + atempo->window / 2) *
706  atempo->tempo;
707 
708  const double ideal_output_position =
709  (double)(prev->position[0] - atempo->origin[0] + atempo->window / 2);
710 
711  const int drift = (int)(prev_output_position - ideal_output_position);
712 
713  const int delta_max = atempo->window / 2;
714  const int correction = yae_align(frag,
715  prev,
716  atempo->window,
717  delta_max,
718  drift,
719  atempo->correlation,
720  atempo->complex_to_real);
721 
722  if (correction) {
723  // adjust fragment position:
724  frag->position[0] -= correction;
725 
726  // clear so that the fragment can be reloaded:
727  frag->nsamples = 0;
728  }
729 
730  return correction;
731 }
732 
733 /**
734  * A helper macro for blending the overlap region of previous
735  * and current audio fragment.
736  */
737 #define yae_blend(scalar_type) \
738  do { \
739  const scalar_type *aaa = (const scalar_type *)a; \
740  const scalar_type *bbb = (const scalar_type *)b; \
741  \
742  scalar_type *out = (scalar_type *)dst; \
743  scalar_type *out_end = (scalar_type *)dst_end; \
744  int64_t i; \
745  \
746  for (i = 0; i < overlap && out < out_end; \
747  i++, atempo->position[1]++, wa++, wb++) { \
748  float w0 = *wa; \
749  float w1 = *wb; \
750  int j; \
751  \
752  for (j = 0; j < atempo->channels; \
753  j++, aaa++, bbb++, out++) { \
754  float t0 = (float)*aaa; \
755  float t1 = (float)*bbb; \
756  \
757  *out = \
758  frag->position[0] + i < 0 ? \
759  *aaa : \
760  (scalar_type)(t0 * w0 + t1 * w1); \
761  } \
762  } \
763  dst = (uint8_t *)out; \
764  } while (0)
765 
766 /**
767  * Blend the overlap region of previous and current audio fragment
768  * and output the results to the given destination buffer.
769  *
770  * @return
771  * 0 if the overlap region was completely stored in the dst buffer,
772  * AVERROR(EAGAIN) if more destination buffer space is required.
773  */
774 static int yae_overlap_add(ATempoContext *atempo,
775  uint8_t **dst_ref,
776  uint8_t *dst_end)
777 {
778  // shortcuts:
779  const AudioFragment *prev = yae_prev_frag(atempo);
780  const AudioFragment *frag = yae_curr_frag(atempo);
781 
782  const int64_t start_here = FFMAX(atempo->position[1],
783  frag->position[1]);
784 
785  const int64_t stop_here = FFMIN(prev->position[1] + prev->nsamples,
786  frag->position[1] + frag->nsamples);
787 
788  const int64_t overlap = stop_here - start_here;
789 
790  const int64_t ia = start_here - prev->position[1];
791  const int64_t ib = start_here - frag->position[1];
792 
793  const float *wa = atempo->hann + ia;
794  const float *wb = atempo->hann + ib;
795 
796  const uint8_t *a = prev->data + ia * atempo->stride;
797  const uint8_t *b = frag->data + ib * atempo->stride;
798 
799  uint8_t *dst = *dst_ref;
800 
801  av_assert0(start_here <= stop_here &&
802  frag->position[1] <= start_here &&
803  overlap <= frag->nsamples);
804 
805  if (atempo->format == AV_SAMPLE_FMT_U8) {
807  } else if (atempo->format == AV_SAMPLE_FMT_S16) {
808  yae_blend(int16_t);
809  } else if (atempo->format == AV_SAMPLE_FMT_S32) {
810  yae_blend(int);
811  } else if (atempo->format == AV_SAMPLE_FMT_FLT) {
812  yae_blend(float);
813  } else if (atempo->format == AV_SAMPLE_FMT_DBL) {
814  yae_blend(double);
815  }
816 
817  // pass-back the updated destination buffer pointer:
818  *dst_ref = dst;
819 
820  return atempo->position[1] == stop_here ? 0 : AVERROR(EAGAIN);
821 }
822 
823 /**
824  * Feed as much data to the filter as it is able to consume
825  * and receive as much processed data in the destination buffer
826  * as it is able to produce or store.
827  */
828 static void
830  const uint8_t **src_ref,
831  const uint8_t *src_end,
832  uint8_t **dst_ref,
833  uint8_t *dst_end)
834 {
835  while (1) {
836  if (atempo->state == YAE_LOAD_FRAGMENT) {
837  // load additional data for the current fragment:
838  if (yae_load_frag(atempo, src_ref, src_end) != 0) {
839  break;
840  }
841 
842  // down-mix to mono:
843  yae_downmix(atempo, yae_curr_frag(atempo));
844 
845  // apply rDFT:
846  av_rdft_calc(atempo->real_to_complex, yae_curr_frag(atempo)->xdat);
847 
848  // must load the second fragment before alignment can start:
849  if (!atempo->nfrag) {
850  yae_advance_to_next_frag(atempo);
851  continue;
852  }
853 
854  atempo->state = YAE_ADJUST_POSITION;
855  }
856 
857  if (atempo->state == YAE_ADJUST_POSITION) {
858  // adjust position for better alignment:
859  if (yae_adjust_position(atempo)) {
860  // reload the fragment at the corrected position, so that the
861  // Hann window blending would not require normalization:
862  atempo->state = YAE_RELOAD_FRAGMENT;
863  } else {
864  atempo->state = YAE_OUTPUT_OVERLAP_ADD;
865  }
866  }
867 
868  if (atempo->state == YAE_RELOAD_FRAGMENT) {
869  // load additional data if necessary due to position adjustment:
870  if (yae_load_frag(atempo, src_ref, src_end) != 0) {
871  break;
872  }
873 
874  // down-mix to mono:
875  yae_downmix(atempo, yae_curr_frag(atempo));
876 
877  // apply rDFT:
878  av_rdft_calc(atempo->real_to_complex, yae_curr_frag(atempo)->xdat);
879 
880  atempo->state = YAE_OUTPUT_OVERLAP_ADD;
881  }
882 
883  if (atempo->state == YAE_OUTPUT_OVERLAP_ADD) {
884  // overlap-add and output the result:
885  if (yae_overlap_add(atempo, dst_ref, dst_end) != 0) {
886  break;
887  }
888 
889  // advance to the next fragment, repeat:
890  yae_advance_to_next_frag(atempo);
891  atempo->state = YAE_LOAD_FRAGMENT;
892  }
893  }
894 }
895 
896 /**
897  * Flush any buffered data from the filter.
898  *
899  * @return
900  * 0 if all data was completely stored in the dst buffer,
901  * AVERROR(EAGAIN) if more destination buffer space is required.
902  */
903 static int yae_flush(ATempoContext *atempo,
904  uint8_t **dst_ref,
905  uint8_t *dst_end)
906 {
907  AudioFragment *frag = yae_curr_frag(atempo);
908  int64_t overlap_end;
909  int64_t start_here;
910  int64_t stop_here;
911  int64_t offset;
912 
913  const uint8_t *src;
914  uint8_t *dst;
915 
916  int src_size;
917  int dst_size;
918  int nbytes;
919 
920  atempo->state = YAE_FLUSH_OUTPUT;
921 
922  if (!atempo->nfrag) {
923  // there is nothing to flush:
924  return 0;
925  }
926 
927  if (atempo->position[0] == frag->position[0] + frag->nsamples &&
928  atempo->position[1] == frag->position[1] + frag->nsamples) {
929  // the current fragment is already flushed:
930  return 0;
931  }
932 
933  if (frag->position[0] + frag->nsamples < atempo->position[0]) {
934  // finish loading the current (possibly partial) fragment:
935  yae_load_frag(atempo, NULL, NULL);
936 
937  if (atempo->nfrag) {
938  // down-mix to mono:
939  yae_downmix(atempo, frag);
940 
941  // apply rDFT:
942  av_rdft_calc(atempo->real_to_complex, frag->xdat);
943 
944  // align current fragment to previous fragment:
945  if (yae_adjust_position(atempo)) {
946  // reload the current fragment due to adjusted position:
947  yae_load_frag(atempo, NULL, NULL);
948  }
949  }
950  }
951 
952  // flush the overlap region:
953  overlap_end = frag->position[1] + FFMIN(atempo->window / 2,
954  frag->nsamples);
955 
956  while (atempo->position[1] < overlap_end) {
957  if (yae_overlap_add(atempo, dst_ref, dst_end) != 0) {
958  return AVERROR(EAGAIN);
959  }
960  }
961 
962  // check whether all of the input samples have been consumed:
963  if (frag->position[0] + frag->nsamples < atempo->position[0]) {
964  yae_advance_to_next_frag(atempo);
965  return AVERROR(EAGAIN);
966  }
967 
968  // flush the remainder of the current fragment:
969  start_here = FFMAX(atempo->position[1], overlap_end);
970  stop_here = frag->position[1] + frag->nsamples;
971  offset = start_here - frag->position[1];
972  av_assert0(start_here <= stop_here && frag->position[1] <= start_here);
973 
974  src = frag->data + offset * atempo->stride;
975  dst = (uint8_t *)*dst_ref;
976 
977  src_size = (int)(stop_here - start_here) * atempo->stride;
978  dst_size = dst_end - dst;
979  nbytes = FFMIN(src_size, dst_size);
980 
981  memcpy(dst, src, nbytes);
982  dst += nbytes;
983 
984  atempo->position[1] += (nbytes / atempo->stride);
985 
986  // pass-back the updated destination buffer pointer:
987  *dst_ref = (uint8_t *)dst;
988 
989  return atempo->position[1] == stop_here ? 0 : AVERROR(EAGAIN);
990 }
991 
993 {
994  ATempoContext *atempo = ctx->priv;
995  atempo->format = AV_SAMPLE_FMT_NONE;
996  atempo->state = YAE_LOAD_FRAGMENT;
997  return 0;
998 }
999 
1001 {
1002  ATempoContext *atempo = ctx->priv;
1003  yae_release_buffers(atempo);
1004 }
1005 
1007 {
1010 
1011  // WSOLA necessitates an internal sliding window ring buffer
1012  // for incoming audio stream.
1013  //
1014  // Planar sample formats are too cumbersome to store in a ring buffer,
1015  // therefore planar sample formats are not supported.
1016  //
1017  static const enum AVSampleFormat sample_fmts[] = {
1024  };
1025  int ret;
1026 
1027  layouts = ff_all_channel_counts();
1028  if (!layouts) {
1029  return AVERROR(ENOMEM);
1030  }
1031  ret = ff_set_common_channel_layouts(ctx, layouts);
1032  if (ret < 0)
1033  return ret;
1034 
1035  formats = ff_make_format_list(sample_fmts);
1036  if (!formats) {
1037  return AVERROR(ENOMEM);
1038  }
1039  ret = ff_set_common_formats(ctx, formats);
1040  if (ret < 0)
1041  return ret;
1042 
1043  formats = ff_all_samplerates();
1044  if (!formats) {
1045  return AVERROR(ENOMEM);
1046  }
1047  return ff_set_common_samplerates(ctx, formats);
1048 }
1049 
1050 static int config_props(AVFilterLink *inlink)
1051 {
1052  AVFilterContext *ctx = inlink->dst;
1053  ATempoContext *atempo = ctx->priv;
1054 
1055  enum AVSampleFormat format = inlink->format;
1056  int sample_rate = (int)inlink->sample_rate;
1057 
1058  return yae_reset(atempo, format, sample_rate, inlink->channels);
1059 }
1060 
1061 static int push_samples(ATempoContext *atempo,
1062  AVFilterLink *outlink,
1063  int n_out)
1064 {
1065  int ret;
1066 
1067  atempo->dst_buffer->sample_rate = outlink->sample_rate;
1068  atempo->dst_buffer->nb_samples = n_out;
1069 
1070  // adjust the PTS:
1071  atempo->dst_buffer->pts =
1072  av_rescale_q(atempo->nsamples_out,
1073  (AVRational){ 1, outlink->sample_rate },
1074  outlink->time_base);
1075 
1076  ret = ff_filter_frame(outlink, atempo->dst_buffer);
1077  atempo->dst_buffer = NULL;
1078  atempo->dst = NULL;
1079  atempo->dst_end = NULL;
1080  if (ret < 0)
1081  return ret;
1082 
1083  atempo->nsamples_out += n_out;
1084  return 0;
1085 }
1086 
1087 static int filter_frame(AVFilterLink *inlink, AVFrame *src_buffer)
1088 {
1089  AVFilterContext *ctx = inlink->dst;
1090  ATempoContext *atempo = ctx->priv;
1091  AVFilterLink *outlink = ctx->outputs[0];
1092 
1093  int ret = 0;
1094  int n_in = src_buffer->nb_samples;
1095  int n_out = (int)(0.5 + ((double)n_in) / atempo->tempo);
1096 
1097  const uint8_t *src = src_buffer->data[0];
1098  const uint8_t *src_end = src + n_in * atempo->stride;
1099 
1100  while (src < src_end) {
1101  if (!atempo->dst_buffer) {
1102  atempo->dst_buffer = ff_get_audio_buffer(outlink, n_out);
1103  if (!atempo->dst_buffer) {
1104  av_frame_free(&src_buffer);
1105  return AVERROR(ENOMEM);
1106  }
1107  av_frame_copy_props(atempo->dst_buffer, src_buffer);
1108 
1109  atempo->dst = atempo->dst_buffer->data[0];
1110  atempo->dst_end = atempo->dst + n_out * atempo->stride;
1111  }
1112 
1113  yae_apply(atempo, &src, src_end, &atempo->dst, atempo->dst_end);
1114 
1115  if (atempo->dst == atempo->dst_end) {
1116  int n_samples = ((atempo->dst - atempo->dst_buffer->data[0]) /
1117  atempo->stride);
1118  ret = push_samples(atempo, outlink, n_samples);
1119  if (ret < 0)
1120  goto end;
1121  }
1122  }
1123 
1124  atempo->nsamples_in += n_in;
1125 end:
1126  av_frame_free(&src_buffer);
1127  return ret;
1128 }
1129 
1130 static int request_frame(AVFilterLink *outlink)
1131 {
1132  AVFilterContext *ctx = outlink->src;
1133  ATempoContext *atempo = ctx->priv;
1134  int ret;
1135 
1136  ret = ff_request_frame(ctx->inputs[0]);
1137 
1138  if (ret == AVERROR_EOF) {
1139  // flush the filter:
1140  int n_max = atempo->ring;
1141  int n_out;
1142  int err = AVERROR(EAGAIN);
1143 
1144  while (err == AVERROR(EAGAIN)) {
1145  if (!atempo->dst_buffer) {
1146  atempo->dst_buffer = ff_get_audio_buffer(outlink, n_max);
1147  if (!atempo->dst_buffer)
1148  return AVERROR(ENOMEM);
1149 
1150  atempo->dst = atempo->dst_buffer->data[0];
1151  atempo->dst_end = atempo->dst + n_max * atempo->stride;
1152  }
1153 
1154  err = yae_flush(atempo, &atempo->dst, atempo->dst_end);
1155 
1156  n_out = ((atempo->dst - atempo->dst_buffer->data[0]) /
1157  atempo->stride);
1158 
1159  if (n_out) {
1160  ret = push_samples(atempo, outlink, n_out);
1161  if (ret < 0)
1162  return ret;
1163  }
1164  }
1165 
1166  av_frame_free(&atempo->dst_buffer);
1167  atempo->dst = NULL;
1168  atempo->dst_end = NULL;
1169 
1170  return AVERROR_EOF;
1171  }
1172 
1173  return ret;
1174 }
1175 
1177  const char *cmd,
1178  const char *arg,
1179  char *res,
1180  int res_len,
1181  int flags)
1182 {
1183  return !strcmp(cmd, "tempo") ? yae_set_tempo(ctx, arg) : AVERROR(ENOSYS);
1184 }
1185 
1186 static const AVFilterPad atempo_inputs[] = {
1187  {
1188  .name = "default",
1189  .type = AVMEDIA_TYPE_AUDIO,
1190  .filter_frame = filter_frame,
1191  .config_props = config_props,
1192  },
1193  { NULL }
1194 };
1195 
1196 static const AVFilterPad atempo_outputs[] = {
1197  {
1198  .name = "default",
1199  .request_frame = request_frame,
1200  .type = AVMEDIA_TYPE_AUDIO,
1201  },
1202  { NULL }
1203 };
1204 
1206  .name = "atempo",
1207  .description = NULL_IF_CONFIG_SMALL("Adjust audio tempo."),
1208  .init = init,
1209  .uninit = uninit,
1210  .query_formats = query_formats,
1211  .process_command = process_command,
1212  .priv_size = sizeof(ATempoContext),
1213  .priv_class = &atempo_class,
1214  .inputs = atempo_inputs,
1215  .outputs = atempo_outputs,
1216 };
#define RE_MALLOC_OR_FAIL(field, field_size)
Definition: af_atempo.c:243
#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
static int push_samples(ATempoContext *atempo, AVFilterLink *outlink, int n_out)
Definition: af_atempo.c:1061
static const char * format[]
Definition: af_aiir.c:330
static void yae_xcorr_via_rdft(FFTSample *xcorr, RDFTContext *complex_to_real, const FFTComplex *xa, const FFTComplex *xb, const int window)
Calculate cross-correlation via rDFT.
Definition: af_atempo.c:611
This structure describes decoded (raw) audio or video data.
Definition: frame.h:226
int64_t origin[2]
Definition: af_atempo.c:128
RDFTContext * complex_to_real
Definition: af_atempo.c:141
FilterState
Filter state machine states.
Definition: af_atempo.c:76
AVOption.
Definition: opt.h:246
RDFTContext * real_to_complex
Definition: af_atempo.c:140
static int yae_set_tempo(AVFilterContext *ctx, const char *arg_tempo)
Definition: af_atempo.c:327
static int config_props(AVFilterLink *inlink)
Definition: af_atempo.c:1050
AVFrame * dst_buffer
Definition: af_atempo.c:145
Main libavfilter public API header.
enum AVSampleFormat format
Definition: af_atempo.c:107
channels
Definition: aptx.c:30
#define AV_OPT_FLAG_AUDIO_PARAM
Definition: opt.h:278
const char * b
Definition: vf_curves.c:116
int av_log2(unsigned v)
Definition: intmath.c:26
static void yae_apply(ATempoContext *atempo, const uint8_t **src_ref, const uint8_t *src_end, uint8_t **dst_ref, uint8_t *dst_end)
Feed as much data to the filter as it is able to consume and receive as much processed data in the de...
Definition: af_atempo.c:829
static const AVFilterPad atempo_outputs[]
Definition: af_atempo.c:1196
FFTSample re
Definition: avfft.h:38
static int request_frame(AVFilterLink *outlink)
Definition: af_atempo.c:1130
AVFILTER_DEFINE_CLASS(atempo)
float * hann
Definition: af_atempo.c:121
#define src
Definition: vp8dsp.c:254
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:283
static void yae_advance_to_next_frag(ATempoContext *atempo)
Prepare for loading next audio fragment.
Definition: af_atempo.c:589
const char * name
Pad name.
Definition: internal.h:60
AVFilterLink ** inputs
array of pointers to input links
Definition: avfilter.h:346
static AudioFragment * yae_prev_frag(ATempoContext *atempo)
Definition: af_atempo.c:173
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:37
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1080
uint8_t * buffer
Definition: af_atempo.c:92
uint8_t
#define av_cold
Definition: attributes.h:82
AV_SAMPLE_FMT_U8
AVOptions.
static av_cold int end(AVCodecContext *avctx)
Definition: avrndec.c:90
static AudioFragment * yae_curr_frag(ATempoContext *atempo)
Definition: af_atempo.c:168
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:319
#define YAE_ATEMPO_MAX
Definition: af_atempo.c:153
Filter state machine.
Definition: af_atempo.c:87
#define AVERROR_EOF
End of file.
Definition: error.h:55
signed 32 bits
Definition: samplefmt.h:62
static av_cold int init(AVFilterContext *ctx)
Definition: af_atempo.c:992
#define av_log(a,...)
A filter pad used for either input or output.
Definition: internal.h:54
static void yae_clear(ATempoContext *atempo)
Reset filter to initial state, do not deallocate existing local buffers.
Definition: af_atempo.c:181
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
static int yae_load_data(ATempoContext *atempo, const uint8_t **src_ref, const uint8_t *src_end, int64_t stop_here)
Populate the internal data buffer on as-needed basis.
Definition: af_atempo.c:434
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:176
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
#define AV_OPT_FLAG_FILTERING_PARAM
a generic parameter which can be set by the user for filtering
Definition: opt.h:291
#define AVERROR(e)
Definition: error.h:43
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:202
A fragment of audio waveform.
Definition: af_atempo.c:56
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:186
uint64_t nsamples_in
Definition: af_atempo.c:148
void * priv
private data for use by the filter
Definition: avfilter.h:353
int64_t position[2]
Definition: af_atempo.c:60
const char * arg
Definition: jacosubdec.c:66
simple assert() macros that are a bit more flexible than ISO C assert().
Definition: avfft.h:73
uint64_t nfrag
Definition: af_atempo.c:134
static const uint8_t offset[127][2]
Definition: vf_spp.c:92
#define FFMAX(a, b)
Definition: common.h:94
float FFTSample
Definition: avfft.h:35
static int yae_flush(ATempoContext *atempo, uint8_t **dst_ref, uint8_t *dst_end)
Flush any buffered data from the filter.
Definition: af_atempo.c:903
void av_rdft_calc(RDFTContext *s, FFTSample *data)
int64_t position[2]
Definition: af_atempo.c:104
static int yae_reset(ATempoContext *atempo, enum AVSampleFormat format, int sample_rate, int channels)
Prepare filter for processing audio data of given format, sample rate and number of channels...
Definition: af_atempo.c:257
static SDL_Window * window
Definition: ffplay.c:365
static const AVFilterPad atempo_inputs[]
Definition: af_atempo.c:1186
audio channel layout utility functions
static int yae_align(AudioFragment *frag, const AudioFragment *prev, const int window, const int delta_max, const int drift, FFTSample *correlation, RDFTContext *complex_to_real)
Calculate alignment offset for given fragment relative to the previous fragment.
Definition: af_atempo.c:645
#define FFMIN(a, b)
Definition: common.h:96
static int yae_overlap_add(ATempoContext *atempo, uint8_t **dst_ref, uint8_t *dst_end)
Blend the overlap region of previous and current audio fragment and output the results to the given d...
Definition: af_atempo.c:774
uint64_t nsamples_out
Definition: af_atempo.c:149
AVFormatContext * ctx
Definition: movenc.c:48
Definition: avfft.h:72
void av_rdft_end(RDFTContext *s)
static int yae_adjust_position(ATempoContext *atempo)
Adjust current fragment position for better alignment with previous fragment.
Definition: af_atempo.c:699
RDFTContext * av_rdft_init(int nbits, enum RDFTransformType trans)
Set up a real FFT.
static int process_command(AVFilterContext *ctx, const char *cmd, const char *arg, char *res, int res_len, int flags)
Definition: af_atempo.c:1176
static const AVFilterPad inputs[]
Definition: af_acontrast.c:193
#define yae_blend(scalar_type)
A helper macro for blending the overlap region of previous and current audio fragment.
Definition: af_atempo.c:737
static const AVFilterPad outputs[]
Definition: af_acontrast.c:203
#define YAE_ATEMPO_MIN
Definition: af_atempo.c:152
A list of supported channel layouts.
Definition: formats.h:85
static int query_formats(AVFilterContext *ctx)
Definition: af_atempo.c:1006
static void yae_release_buffers(ATempoContext *atempo)
Reset filter to initial state and deallocate all buffers.
Definition: af_atempo.c:221
double av_strtod(const char *numstr, char **tail)
Parse the string in numstr and return its value as a double.
Definition: eval.c:106
sample_rate
AVSampleFormat
Audio sample formats.
Definition: samplefmt.h:58
FFT functions.
FFTSample * correlation
Definition: af_atempo.c:142
static void correlation(int32_t *corr, int32_t *ener, int16_t *buffer, int16_t lag, int16_t blen, int16_t srange, int16_t scale)
Definition: ilbcdec.c:912
static int filter_frame(AVFilterLink *inlink, AVFrame *src_buffer)
Definition: af_atempo.c:1087
Describe the class of an AVClass context structure.
Definition: log.h:67
int sample_rate
Sample rate of the audio data.
Definition: frame.h:399
Filter definition.
Definition: avfilter.h:144
Rational number (pair of numerator and denominator).
Definition: rational.h:58
AudioFragment frag[2]
Definition: af_atempo.c:131
uint8_t * dst_end
Definition: af_atempo.c:147
const char * name
Filter name.
Definition: avfilter.h:148
AVFilter ff_af_atempo
Definition: af_atempo.c:1205
uint8_t * dst
Definition: af_atempo.c:146
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
#define yae_init_xdat(scalar_type, scalar_max)
A helper macro for initializing complex data buffer with scalar data of a given type.
Definition: af_atempo.c:356
FFTSample * xdat
Definition: af_atempo.c:70
#define flags(name, subs,...)
Definition: cbs_av1.c:596
FilterState state
Definition: af_atempo.c:137
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:240
int av_get_bytes_per_sample(enum AVSampleFormat sample_fmt)
Return number of bytes per sample.
Definition: samplefmt.c:106
int
FFTSample im
Definition: avfft.h:38
double tempo
Definition: af_atempo.c:124
signed 16 bits
Definition: samplefmt.h:61
static av_cold void uninit(AVFilterContext *ctx)
Definition: af_atempo.c:1000
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 int yae_load_frag(ATempoContext *atempo, const uint8_t **src_ref, const uint8_t *src_end)
Populate current audio fragment data buffer.
Definition: af_atempo.c:509
static void yae_downmix(ATempoContext *atempo, AudioFragment *frag)
Initialize complex data buffer of a given audio fragment with down-mixed mono data of appropriate sca...
Definition: af_atempo.c:406
static enum AVSampleFormat sample_fmts[]
Definition: adpcmenc.c:701
static const AVOption atempo_options[]
Definition: af_atempo.c:157
#define av_freep(p)
void INT64 start
Definition: avisynth_c.h:690
#define M_PI
Definition: mathematics.h:52
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
AVFilterChannelLayouts * ff_all_channel_counts(void)
Construct an AVFilterChannelLayouts coding for any channel layout, with known or unknown disposition...
Definition: formats.c:410
uint8_t * data
Definition: af_atempo.c:63
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:292
int ff_set_common_samplerates(AVFilterContext *ctx, AVFilterFormats *samplerates)
Definition: formats.c:556
int av_frame_copy_props(AVFrame *dst, const AVFrame *src)
Copy only "metadata" fields from src to dst.
Definition: frame.c:654
#define OFFSET(x)
Definition: af_atempo.c:155
simple arithmetic expression evaluator