FFmpeg
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
transcode_aac.c
Go to the documentation of this file.
1 /*
2  * This file is part of FFmpeg.
3  *
4  * FFmpeg is free software; you can redistribute it and/or
5  * modify it under the terms of the GNU Lesser General Public
6  * License as published by the Free Software Foundation; either
7  * version 2.1 of the License, or (at your option) any later version.
8  *
9  * FFmpeg is distributed in the hope that it will be useful,
10  * but WITHOUT ANY WARRANTY; without even the implied warranty of
11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
12  * Lesser General Public License for more details.
13  *
14  * You should have received a copy of the GNU Lesser General Public
15  * License along with FFmpeg; if not, write to the Free Software
16  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
17  */
18 
19 /**
20  * @file
21  * simple audio converter
22  *
23  * @example transcode_aac.c
24  * Convert an input audio file to AAC in an MP4 container using FFmpeg.
25  * @author Andreas Unterweger (dustsigns@gmail.com)
26  */
27 
28 #include <stdio.h>
29 
30 #include "libavformat/avformat.h"
31 #include "libavformat/avio.h"
32 
33 #include "libavcodec/avcodec.h"
34 
35 #include "libavutil/audio_fifo.h"
36 #include "libavutil/avassert.h"
37 #include "libavutil/avstring.h"
38 #include "libavutil/frame.h"
39 #include "libavutil/opt.h"
40 
42 
43 /** The output bit rate in kbit/s */
44 #define OUTPUT_BIT_RATE 96000
45 /** The number of output channels */
46 #define OUTPUT_CHANNELS 2
47 
48 /**
49  * Convert an error code into a text message.
50  * @param error Error code to be converted
51  * @return Corresponding error text (not thread-safe)
52  */
53 static const char *get_error_text(const int error)
54 {
55  static char error_buffer[255];
56  av_strerror(error, error_buffer, sizeof(error_buffer));
57  return error_buffer;
58 }
59 
60 /** Open an input file and the required decoder. */
61 static int open_input_file(const char *filename,
62  AVFormatContext **input_format_context,
63  AVCodecContext **input_codec_context)
64 {
65  AVCodecContext *avctx;
66  AVCodec *input_codec;
67  int error;
68 
69  /** Open the input file to read from it. */
70  if ((error = avformat_open_input(input_format_context, filename, NULL,
71  NULL)) < 0) {
72  fprintf(stderr, "Could not open input file '%s' (error '%s')\n",
73  filename, get_error_text(error));
74  *input_format_context = NULL;
75  return error;
76  }
77 
78  /** Get information on the input file (number of streams etc.). */
79  if ((error = avformat_find_stream_info(*input_format_context, NULL)) < 0) {
80  fprintf(stderr, "Could not open find stream info (error '%s')\n",
81  get_error_text(error));
82  avformat_close_input(input_format_context);
83  return error;
84  }
85 
86  /** Make sure that there is only one stream in the input file. */
87  if ((*input_format_context)->nb_streams != 1) {
88  fprintf(stderr, "Expected one audio input stream, but found %d\n",
89  (*input_format_context)->nb_streams);
90  avformat_close_input(input_format_context);
91  return AVERROR_EXIT;
92  }
93 
94  /** Find a decoder for the audio stream. */
95  if (!(input_codec = avcodec_find_decoder((*input_format_context)->streams[0]->codecpar->codec_id))) {
96  fprintf(stderr, "Could not find input codec\n");
97  avformat_close_input(input_format_context);
98  return AVERROR_EXIT;
99  }
100 
101  /** allocate a new decoding context */
102  avctx = avcodec_alloc_context3(input_codec);
103  if (!avctx) {
104  fprintf(stderr, "Could not allocate a decoding context\n");
105  avformat_close_input(input_format_context);
106  return AVERROR(ENOMEM);
107  }
108 
109  /** initialize the stream parameters with demuxer information */
110  error = avcodec_parameters_to_context(avctx, (*input_format_context)->streams[0]->codecpar);
111  if (error < 0) {
112  avformat_close_input(input_format_context);
113  avcodec_free_context(&avctx);
114  return error;
115  }
116 
117  /** Open the decoder for the audio stream to use it later. */
118  if ((error = avcodec_open2(avctx, input_codec, NULL)) < 0) {
119  fprintf(stderr, "Could not open input codec (error '%s')\n",
120  get_error_text(error));
121  avcodec_free_context(&avctx);
122  avformat_close_input(input_format_context);
123  return error;
124  }
125 
126  /** Save the decoder context for easier access later. */
127  *input_codec_context = avctx;
128 
129  return 0;
130 }
131 
132 /**
133  * Open an output file and the required encoder.
134  * Also set some basic encoder parameters.
135  * Some of these parameters are based on the input file's parameters.
136  */
137 static int open_output_file(const char *filename,
138  AVCodecContext *input_codec_context,
139  AVFormatContext **output_format_context,
140  AVCodecContext **output_codec_context)
141 {
142  AVCodecContext *avctx = NULL;
143  AVIOContext *output_io_context = NULL;
144  AVStream *stream = NULL;
145  AVCodec *output_codec = NULL;
146  int error;
147 
148  /** Open the output file to write to it. */
149  if ((error = avio_open(&output_io_context, filename,
150  AVIO_FLAG_WRITE)) < 0) {
151  fprintf(stderr, "Could not open output file '%s' (error '%s')\n",
152  filename, get_error_text(error));
153  return error;
154  }
155 
156  /** Create a new format context for the output container format. */
157  if (!(*output_format_context = avformat_alloc_context())) {
158  fprintf(stderr, "Could not allocate output format context\n");
159  return AVERROR(ENOMEM);
160  }
161 
162  /** Associate the output file (pointer) with the container format context. */
163  (*output_format_context)->pb = output_io_context;
164 
165  /** Guess the desired container format based on the file extension. */
166  if (!((*output_format_context)->oformat = av_guess_format(NULL, filename,
167  NULL))) {
168  fprintf(stderr, "Could not find output file format\n");
169  goto cleanup;
170  }
171 
172  av_strlcpy((*output_format_context)->filename, filename,
173  sizeof((*output_format_context)->filename));
174 
175  /** Find the encoder to be used by its name. */
176  if (!(output_codec = avcodec_find_encoder(AV_CODEC_ID_AAC))) {
177  fprintf(stderr, "Could not find an AAC encoder.\n");
178  goto cleanup;
179  }
180 
181  /** Create a new audio stream in the output file container. */
182  if (!(stream = avformat_new_stream(*output_format_context, NULL))) {
183  fprintf(stderr, "Could not create new stream\n");
184  error = AVERROR(ENOMEM);
185  goto cleanup;
186  }
187 
188  avctx = avcodec_alloc_context3(output_codec);
189  if (!avctx) {
190  fprintf(stderr, "Could not allocate an encoding context\n");
191  error = AVERROR(ENOMEM);
192  goto cleanup;
193  }
194 
195  /**
196  * Set the basic encoder parameters.
197  * The input file's sample rate is used to avoid a sample rate conversion.
198  */
199  avctx->channels = OUTPUT_CHANNELS;
201  avctx->sample_rate = input_codec_context->sample_rate;
202  avctx->sample_fmt = output_codec->sample_fmts[0];
203  avctx->bit_rate = OUTPUT_BIT_RATE;
204 
205  /** Allow the use of the experimental AAC encoder */
207 
208  /** Set the sample rate for the container. */
209  stream->time_base.den = input_codec_context->sample_rate;
210  stream->time_base.num = 1;
211 
212  /**
213  * Some container formats (like MP4) require global headers to be present
214  * Mark the encoder so that it behaves accordingly.
215  */
216  if ((*output_format_context)->oformat->flags & AVFMT_GLOBALHEADER)
218 
219  /** Open the encoder for the audio stream to use it later. */
220  if ((error = avcodec_open2(avctx, output_codec, NULL)) < 0) {
221  fprintf(stderr, "Could not open output codec (error '%s')\n",
222  get_error_text(error));
223  goto cleanup;
224  }
225 
226  error = avcodec_parameters_from_context(stream->codecpar, avctx);
227  if (error < 0) {
228  fprintf(stderr, "Could not initialize stream parameters\n");
229  goto cleanup;
230  }
231 
232  /** Save the encoder context for easier access later. */
233  *output_codec_context = avctx;
234 
235  return 0;
236 
237 cleanup:
238  avcodec_free_context(&avctx);
239  avio_closep(&(*output_format_context)->pb);
240  avformat_free_context(*output_format_context);
241  *output_format_context = NULL;
242  return error < 0 ? error : AVERROR_EXIT;
243 }
244 
245 /** Initialize one data packet for reading or writing. */
246 static void init_packet(AVPacket *packet)
247 {
248  av_init_packet(packet);
249  /** Set the packet data and size so that it is recognized as being empty. */
250  packet->data = NULL;
251  packet->size = 0;
252 }
253 
254 /** Initialize one audio frame for reading from the input file */
256 {
257  if (!(*frame = av_frame_alloc())) {
258  fprintf(stderr, "Could not allocate input frame\n");
259  return AVERROR(ENOMEM);
260  }
261  return 0;
262 }
263 
264 /**
265  * Initialize the audio resampler based on the input and output codec settings.
266  * If the input and output sample formats differ, a conversion is required
267  * libswresample takes care of this, but requires initialization.
268  */
269 static int init_resampler(AVCodecContext *input_codec_context,
270  AVCodecContext *output_codec_context,
271  SwrContext **resample_context)
272 {
273  int error;
274 
275  /**
276  * Create a resampler context for the conversion.
277  * Set the conversion parameters.
278  * Default channel layouts based on the number of channels
279  * are assumed for simplicity (they are sometimes not detected
280  * properly by the demuxer and/or decoder).
281  */
282  *resample_context = swr_alloc_set_opts(NULL,
283  av_get_default_channel_layout(output_codec_context->channels),
284  output_codec_context->sample_fmt,
285  output_codec_context->sample_rate,
286  av_get_default_channel_layout(input_codec_context->channels),
287  input_codec_context->sample_fmt,
288  input_codec_context->sample_rate,
289  0, NULL);
290  if (!*resample_context) {
291  fprintf(stderr, "Could not allocate resample context\n");
292  return AVERROR(ENOMEM);
293  }
294  /**
295  * Perform a sanity check so that the number of converted samples is
296  * not greater than the number of samples to be converted.
297  * If the sample rates differ, this case has to be handled differently
298  */
299  av_assert0(output_codec_context->sample_rate == input_codec_context->sample_rate);
300 
301  /** Open the resampler with the specified parameters. */
302  if ((error = swr_init(*resample_context)) < 0) {
303  fprintf(stderr, "Could not open resample context\n");
304  swr_free(resample_context);
305  return error;
306  }
307  return 0;
308 }
309 
310 /** Initialize a FIFO buffer for the audio samples to be encoded. */
311 static int init_fifo(AVAudioFifo **fifo, AVCodecContext *output_codec_context)
312 {
313  /** Create the FIFO buffer based on the specified output sample format. */
314  if (!(*fifo = av_audio_fifo_alloc(output_codec_context->sample_fmt,
315  output_codec_context->channels, 1))) {
316  fprintf(stderr, "Could not allocate FIFO\n");
317  return AVERROR(ENOMEM);
318  }
319  return 0;
320 }
321 
322 /** Write the header of the output file container. */
323 static int write_output_file_header(AVFormatContext *output_format_context)
324 {
325  int error;
326  if ((error = avformat_write_header(output_format_context, NULL)) < 0) {
327  fprintf(stderr, "Could not write output file header (error '%s')\n",
328  get_error_text(error));
329  return error;
330  }
331  return 0;
332 }
333 
334 /** Decode one audio frame from the input file. */
336  AVFormatContext *input_format_context,
337  AVCodecContext *input_codec_context,
338  int *data_present, int *finished)
339 {
340  /** Packet used for temporary storage. */
341  AVPacket input_packet;
342  int error;
343  init_packet(&input_packet);
344 
345  /** Read one audio frame from the input file into a temporary packet. */
346  if ((error = av_read_frame(input_format_context, &input_packet)) < 0) {
347  /** If we are at the end of the file, flush the decoder below. */
348  if (error == AVERROR_EOF)
349  *finished = 1;
350  else {
351  fprintf(stderr, "Could not read frame (error '%s')\n",
352  get_error_text(error));
353  return error;
354  }
355  }
356 
357  /**
358  * Decode the audio frame stored in the temporary packet.
359  * The input audio stream decoder is used to do this.
360  * If we are at the end of the file, pass an empty packet to the decoder
361  * to flush it.
362  */
363  if ((error = avcodec_decode_audio4(input_codec_context, frame,
364  data_present, &input_packet)) < 0) {
365  fprintf(stderr, "Could not decode frame (error '%s')\n",
366  get_error_text(error));
367  av_packet_unref(&input_packet);
368  return error;
369  }
370 
371  /**
372  * If the decoder has not been flushed completely, we are not finished,
373  * so that this function has to be called again.
374  */
375  if (*finished && *data_present)
376  *finished = 0;
377  av_packet_unref(&input_packet);
378  return 0;
379 }
380 
381 /**
382  * Initialize a temporary storage for the specified number of audio samples.
383  * The conversion requires temporary storage due to the different format.
384  * The number of audio samples to be allocated is specified in frame_size.
385  */
386 static int init_converted_samples(uint8_t ***converted_input_samples,
387  AVCodecContext *output_codec_context,
388  int frame_size)
389 {
390  int error;
391 
392  /**
393  * Allocate as many pointers as there are audio channels.
394  * Each pointer will later point to the audio samples of the corresponding
395  * channels (although it may be NULL for interleaved formats).
396  */
397  if (!(*converted_input_samples = calloc(output_codec_context->channels,
398  sizeof(**converted_input_samples)))) {
399  fprintf(stderr, "Could not allocate converted input sample pointers\n");
400  return AVERROR(ENOMEM);
401  }
402 
403  /**
404  * Allocate memory for the samples of all channels in one consecutive
405  * block for convenience.
406  */
407  if ((error = av_samples_alloc(*converted_input_samples, NULL,
408  output_codec_context->channels,
409  frame_size,
410  output_codec_context->sample_fmt, 0)) < 0) {
411  fprintf(stderr,
412  "Could not allocate converted input samples (error '%s')\n",
413  get_error_text(error));
414  av_freep(&(*converted_input_samples)[0]);
415  free(*converted_input_samples);
416  return error;
417  }
418  return 0;
419 }
420 
421 /**
422  * Convert the input audio samples into the output sample format.
423  * The conversion happens on a per-frame basis, the size of which is specified
424  * by frame_size.
425  */
426 static int convert_samples(const uint8_t **input_data,
427  uint8_t **converted_data, const int frame_size,
428  SwrContext *resample_context)
429 {
430  int error;
431 
432  /** Convert the samples using the resampler. */
433  if ((error = swr_convert(resample_context,
434  converted_data, frame_size,
435  input_data , frame_size)) < 0) {
436  fprintf(stderr, "Could not convert input samples (error '%s')\n",
437  get_error_text(error));
438  return error;
439  }
440 
441  return 0;
442 }
443 
444 /** Add converted input audio samples to the FIFO buffer for later processing. */
446  uint8_t **converted_input_samples,
447  const int frame_size)
448 {
449  int error;
450 
451  /**
452  * Make the FIFO as large as it needs to be to hold both,
453  * the old and the new samples.
454  */
455  if ((error = av_audio_fifo_realloc(fifo, av_audio_fifo_size(fifo) + frame_size)) < 0) {
456  fprintf(stderr, "Could not reallocate FIFO\n");
457  return error;
458  }
459 
460  /** Store the new samples in the FIFO buffer. */
461  if (av_audio_fifo_write(fifo, (void **)converted_input_samples,
462  frame_size) < frame_size) {
463  fprintf(stderr, "Could not write data to FIFO\n");
464  return AVERROR_EXIT;
465  }
466  return 0;
467 }
468 
469 /**
470  * Read one audio frame from the input file, decodes, converts and stores
471  * it in the FIFO buffer.
472  */
474  AVFormatContext *input_format_context,
475  AVCodecContext *input_codec_context,
476  AVCodecContext *output_codec_context,
477  SwrContext *resampler_context,
478  int *finished)
479 {
480  /** Temporary storage of the input samples of the frame read from the file. */
481  AVFrame *input_frame = NULL;
482  /** Temporary storage for the converted input samples. */
483  uint8_t **converted_input_samples = NULL;
484  int data_present;
485  int ret = AVERROR_EXIT;
486 
487  /** Initialize temporary storage for one input frame. */
488  if (init_input_frame(&input_frame))
489  goto cleanup;
490  /** Decode one frame worth of audio samples. */
491  if (decode_audio_frame(input_frame, input_format_context,
492  input_codec_context, &data_present, finished))
493  goto cleanup;
494  /**
495  * If we are at the end of the file and there are no more samples
496  * in the decoder which are delayed, we are actually finished.
497  * This must not be treated as an error.
498  */
499  if (*finished && !data_present) {
500  ret = 0;
501  goto cleanup;
502  }
503  /** If there is decoded data, convert and store it */
504  if (data_present) {
505  /** Initialize the temporary storage for the converted input samples. */
506  if (init_converted_samples(&converted_input_samples, output_codec_context,
507  input_frame->nb_samples))
508  goto cleanup;
509 
510  /**
511  * Convert the input samples to the desired output sample format.
512  * This requires a temporary storage provided by converted_input_samples.
513  */
514  if (convert_samples((const uint8_t**)input_frame->extended_data, converted_input_samples,
515  input_frame->nb_samples, resampler_context))
516  goto cleanup;
517 
518  /** Add the converted input samples to the FIFO buffer for later processing. */
519  if (add_samples_to_fifo(fifo, converted_input_samples,
520  input_frame->nb_samples))
521  goto cleanup;
522  ret = 0;
523  }
524  ret = 0;
525 
526 cleanup:
527  if (converted_input_samples) {
528  av_freep(&converted_input_samples[0]);
529  free(converted_input_samples);
530  }
531  av_frame_free(&input_frame);
532 
533  return ret;
534 }
535 
536 /**
537  * Initialize one input frame for writing to the output file.
538  * The frame will be exactly frame_size samples large.
539  */
541  AVCodecContext *output_codec_context,
542  int frame_size)
543 {
544  int error;
545 
546  /** Create a new frame to store the audio samples. */
547  if (!(*frame = av_frame_alloc())) {
548  fprintf(stderr, "Could not allocate output frame\n");
549  return AVERROR_EXIT;
550  }
551 
552  /**
553  * Set the frame's parameters, especially its size and format.
554  * av_frame_get_buffer needs this to allocate memory for the
555  * audio samples of the frame.
556  * Default channel layouts based on the number of channels
557  * are assumed for simplicity.
558  */
559  (*frame)->nb_samples = frame_size;
560  (*frame)->channel_layout = output_codec_context->channel_layout;
561  (*frame)->format = output_codec_context->sample_fmt;
562  (*frame)->sample_rate = output_codec_context->sample_rate;
563 
564  /**
565  * Allocate the samples of the created frame. This call will make
566  * sure that the audio frame can hold as many samples as specified.
567  */
568  if ((error = av_frame_get_buffer(*frame, 0)) < 0) {
569  fprintf(stderr, "Could allocate output frame samples (error '%s')\n",
570  get_error_text(error));
571  av_frame_free(frame);
572  return error;
573  }
574 
575  return 0;
576 }
577 
578 /** Global timestamp for the audio frames */
579 static int64_t pts = 0;
580 
581 /** Encode one frame worth of audio to the output file. */
583  AVFormatContext *output_format_context,
584  AVCodecContext *output_codec_context,
585  int *data_present)
586 {
587  /** Packet used for temporary storage. */
589  int error;
590  init_packet(&output_packet);
591 
592  /** Set a timestamp based on the sample rate for the container. */
593  if (frame) {
594  frame->pts = pts;
595  pts += frame->nb_samples;
596  }
597 
598  /**
599  * Encode the audio frame and store it in the temporary packet.
600  * The output audio stream encoder is used to do this.
601  */
602  if ((error = avcodec_encode_audio2(output_codec_context, &output_packet,
603  frame, data_present)) < 0) {
604  fprintf(stderr, "Could not encode frame (error '%s')\n",
605  get_error_text(error));
606  av_packet_unref(&output_packet);
607  return error;
608  }
609 
610  /** Write one audio frame from the temporary packet to the output file. */
611  if (*data_present) {
612  if ((error = av_write_frame(output_format_context, &output_packet)) < 0) {
613  fprintf(stderr, "Could not write frame (error '%s')\n",
614  get_error_text(error));
615  av_packet_unref(&output_packet);
616  return error;
617  }
618 
619  av_packet_unref(&output_packet);
620  }
621 
622  return 0;
623 }
624 
625 /**
626  * Load one audio frame from the FIFO buffer, encode and write it to the
627  * output file.
628  */
630  AVFormatContext *output_format_context,
631  AVCodecContext *output_codec_context)
632 {
633  /** Temporary storage of the output samples of the frame written to the file. */
635  /**
636  * Use the maximum number of possible samples per frame.
637  * If there is less than the maximum possible frame size in the FIFO
638  * buffer use this number. Otherwise, use the maximum possible frame size
639  */
640  const int frame_size = FFMIN(av_audio_fifo_size(fifo),
641  output_codec_context->frame_size);
642  int data_written;
643 
644  /** Initialize temporary storage for one output frame. */
645  if (init_output_frame(&output_frame, output_codec_context, frame_size))
646  return AVERROR_EXIT;
647 
648  /**
649  * Read as many samples from the FIFO buffer as required to fill the frame.
650  * The samples are stored in the frame temporarily.
651  */
652  if (av_audio_fifo_read(fifo, (void **)output_frame->data, frame_size) < frame_size) {
653  fprintf(stderr, "Could not read data from FIFO\n");
654  av_frame_free(&output_frame);
655  return AVERROR_EXIT;
656  }
657 
658  /** Encode one frame worth of audio samples. */
659  if (encode_audio_frame(output_frame, output_format_context,
660  output_codec_context, &data_written)) {
661  av_frame_free(&output_frame);
662  return AVERROR_EXIT;
663  }
664  av_frame_free(&output_frame);
665  return 0;
666 }
667 
668 /** Write the trailer of the output file container. */
669 static int write_output_file_trailer(AVFormatContext *output_format_context)
670 {
671  int error;
672  if ((error = av_write_trailer(output_format_context)) < 0) {
673  fprintf(stderr, "Could not write output file trailer (error '%s')\n",
674  get_error_text(error));
675  return error;
676  }
677  return 0;
678 }
679 
680 /** Convert an audio file to an AAC file in an MP4 container. */
681 int main(int argc, char **argv)
682 {
683  AVFormatContext *input_format_context = NULL, *output_format_context = NULL;
684  AVCodecContext *input_codec_context = NULL, *output_codec_context = NULL;
685  SwrContext *resample_context = NULL;
686  AVAudioFifo *fifo = NULL;
687  int ret = AVERROR_EXIT;
688 
689  if (argc < 3) {
690  fprintf(stderr, "Usage: %s <input file> <output file>\n", argv[0]);
691  exit(1);
692  }
693 
694  /** Register all codecs and formats so that they can be used. */
695  av_register_all();
696  /** Open the input file for reading. */
697  if (open_input_file(argv[1], &input_format_context,
698  &input_codec_context))
699  goto cleanup;
700  /** Open the output file for writing. */
701  if (open_output_file(argv[2], input_codec_context,
702  &output_format_context, &output_codec_context))
703  goto cleanup;
704  /** Initialize the resampler to be able to convert audio sample formats. */
705  if (init_resampler(input_codec_context, output_codec_context,
706  &resample_context))
707  goto cleanup;
708  /** Initialize the FIFO buffer to store audio samples to be encoded. */
709  if (init_fifo(&fifo, output_codec_context))
710  goto cleanup;
711  /** Write the header of the output file container. */
712  if (write_output_file_header(output_format_context))
713  goto cleanup;
714 
715  /**
716  * Loop as long as we have input samples to read or output samples
717  * to write; abort as soon as we have neither.
718  */
719  while (1) {
720  /** Use the encoder's desired frame size for processing. */
721  const int output_frame_size = output_codec_context->frame_size;
722  int finished = 0;
723 
724  /**
725  * Make sure that there is one frame worth of samples in the FIFO
726  * buffer so that the encoder can do its work.
727  * Since the decoder's and the encoder's frame size may differ, we
728  * need to FIFO buffer to store as many frames worth of input samples
729  * that they make up at least one frame worth of output samples.
730  */
731  while (av_audio_fifo_size(fifo) < output_frame_size) {
732  /**
733  * Decode one frame worth of audio samples, convert it to the
734  * output sample format and put it into the FIFO buffer.
735  */
736  if (read_decode_convert_and_store(fifo, input_format_context,
737  input_codec_context,
738  output_codec_context,
739  resample_context, &finished))
740  goto cleanup;
741 
742  /**
743  * If we are at the end of the input file, we continue
744  * encoding the remaining audio samples to the output file.
745  */
746  if (finished)
747  break;
748  }
749 
750  /**
751  * If we have enough samples for the encoder, we encode them.
752  * At the end of the file, we pass the remaining samples to
753  * the encoder.
754  */
755  while (av_audio_fifo_size(fifo) >= output_frame_size ||
756  (finished && av_audio_fifo_size(fifo) > 0))
757  /**
758  * Take one frame worth of audio samples from the FIFO buffer,
759  * encode it and write it to the output file.
760  */
761  if (load_encode_and_write(fifo, output_format_context,
762  output_codec_context))
763  goto cleanup;
764 
765  /**
766  * If we are at the end of the input file and have encoded
767  * all remaining samples, we can exit this loop and finish.
768  */
769  if (finished) {
770  int data_written;
771  /** Flush the encoder as it may have delayed frames. */
772  do {
773  if (encode_audio_frame(NULL, output_format_context,
774  output_codec_context, &data_written))
775  goto cleanup;
776  } while (data_written);
777  break;
778  }
779  }
780 
781  /** Write the trailer of the output file container. */
782  if (write_output_file_trailer(output_format_context))
783  goto cleanup;
784  ret = 0;
785 
786 cleanup:
787  if (fifo)
788  av_audio_fifo_free(fifo);
789  swr_free(&resample_context);
790  if (output_codec_context)
791  avcodec_free_context(&output_codec_context);
792  if (output_format_context) {
793  avio_closep(&output_format_context->pb);
794  avformat_free_context(output_format_context);
795  }
796  if (input_codec_context)
797  avcodec_free_context(&input_codec_context);
798  if (input_format_context)
799  avformat_close_input(&input_format_context);
800 
801  return ret;
802 }
int avio_open(AVIOContext **s, const char *url, int flags)
Create and initialize a AVIOContext for accessing the resource indicated by url.
Definition: aviobuf.c:1027
#define FF_COMPLIANCE_EXPERIMENTAL
Allow nonstandardized experimental things.
Definition: avcodec.h:2871
#define NULL
Definition: coverity.c:32
Bytestream IO Context.
Definition: avio.h:147
AVAudioFifo * av_audio_fifo_alloc(enum AVSampleFormat sample_fmt, int channels, int nb_samples)
Allocate an AVAudioFifo.
Definition: audio_fifo.c:60
Buffered I/O operations.
int av_audio_fifo_read(AVAudioFifo *af, void **data, int nb_samples)
Read data from an AVAudioFifo.
Definition: audio_fifo.c:182
This structure describes decoded (raw) audio or video data.
Definition: frame.h:184
AVCodec * avcodec_find_encoder(enum AVCodecID id)
Find a registered encoder with a matching codec ID.
Definition: utils.c:3044
int main(int argc, char **argv)
Convert an audio file to an AAC file in an MP4 container.
int av_write_frame(AVFormatContext *s, AVPacket *pkt)
Write a packet to an output media file.
Definition: mux.c:820
int64_t bit_rate
the average bitrate
Definition: avcodec.h:1714
void av_audio_fifo_free(AVAudioFifo *af)
Free an AVAudioFifo.
Definition: audio_fifo.c:45
int num
numerator
Definition: rational.h:44
int size
Definition: avcodec.h:1581
#define AVIO_FLAG_WRITE
write-only
Definition: avio.h:607
attribute_deprecated int avcodec_encode_audio2(AVCodecContext *avctx, AVPacket *avpkt, const AVFrame *frame, int *got_packet_ptr)
Encode a frame of audio.
Definition: utils.c:1776
attribute_deprecated int avcodec_decode_audio4(AVCodecContext *avctx, AVFrame *frame, int *got_frame_ptr, const AVPacket *avpkt)
Decode the audio frame of size avpkt->size from avpkt->data into frame.
Definition: utils.c:2281
AVCodec.
Definition: avcodec.h:3542
static void init_packet(AVPacket *packet)
Initialize one data packet for reading or writing.
static int init_resampler(AVCodecContext *input_codec_context, AVCodecContext *output_codec_context, SwrContext **resample_context)
Initialize the audio resampler based on the input and output codec settings.
static int open_input_file(const char *filename, AVFormatContext **input_format_context, AVCodecContext **input_codec_context)
Open an input file and the required decoder.
Definition: transcode_aac.c:61
Format I/O context.
Definition: avformat.h:1325
static int encode_audio_frame(AVFrame *frame, AVFormatContext *output_format_context, AVCodecContext *output_codec_context, int *data_present)
Encode one frame worth of audio to the output file.
static int output_packet(AVFormatContext *ctx, int flush)
Definition: mpegenc.c:961
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:37
enum AVSampleFormat sample_fmt
audio sample format
Definition: avcodec.h:2418
uint8_t
AVFrame * av_frame_alloc(void)
Allocate an AVFrame and set its fields to default values.
Definition: frame.c:140
AVOptions.
static int read_decode_convert_and_store(AVAudioFifo *fifo, AVFormatContext *input_format_context, AVCodecContext *input_codec_context, AVCodecContext *output_codec_context, SwrContext *resampler_context, int *finished)
Read one audio frame from the input file, decodes, converts and stores it in the FIFO buffer...
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:268
static int convert_samples(const uint8_t **input_data, uint8_t **converted_data, const int frame_size, SwrContext *resample_context)
Convert the input audio samples into the output sample format.
AVStream * avformat_new_stream(AVFormatContext *s, const AVCodec *c)
Add a new stream to a media file.
Definition: utils.c:4065
int avcodec_parameters_to_context(AVCodecContext *codec, const AVCodecParameters *par)
Fill the codec context based on the values from the supplied codec parameters.
Definition: utils.c:4133
static AVFrame * frame
AVFormatContext * avformat_alloc_context(void)
Allocate an AVFormatContext.
Definition: options.c:132
uint8_t * data
Definition: avcodec.h:1580
#define AVERROR_EOF
End of file.
Definition: error.h:55
static int init_input_frame(AVFrame **frame)
Initialize one audio frame for reading from the input file.
libswresample public header
static int write_output_file_header(AVFormatContext *output_format_context)
Write the header of the output file container.
#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:153
The libswresample context.
int flags
AV_CODEC_FLAG_*.
Definition: avcodec.h:1744
simple assert() macros that are a bit more flexible than ISO C assert().
static int load_encode_and_write(AVAudioFifo *fifo, AVFormatContext *output_format_context, AVCodecContext *output_codec_context)
Load one audio frame from the FIFO buffer, encode and write it to the output file.
size_t av_strlcpy(char *dst, const char *src, size_t size)
Copy the string src to dst, but no more than size - 1 bytes, and null-terminate dst.
Definition: avstring.c:83
reference-counted frame API
uint64_t channel_layout
Audio channel layout.
Definition: avcodec.h:2461
Context for an Audio FIFO Buffer.
Definition: audio_fifo.c:34
int av_audio_fifo_size(AVAudioFifo *af)
Get the current number of samples in the AVAudioFifo available for reading.
Definition: audio_fifo.c:229
int av_audio_fifo_realloc(AVAudioFifo *af, int nb_samples)
Reallocate an AVAudioFifo.
Definition: audio_fifo.c:97
av_warn_unused_result int avformat_write_header(AVFormatContext *s, AVDictionary **options)
Allocate the stream private data and write the stream header to an output media file.
Definition: mux.c:496
#define FFMIN(a, b)
Definition: common.h:96
static int open_output_file(const char *filename, AVCodecContext *input_codec_context, AVFormatContext **output_format_context, AVCodecContext **output_codec_context)
Open an output file and the required encoder.
AVCodecContext * avcodec_alloc_context3(const AVCodec *codec)
Allocate an AVCodecContext and set its fields to default values.
Definition: options.c:156
static int decode_audio_frame(AVFrame *frame, AVFormatContext *input_format_context, AVCodecContext *input_codec_context, int *data_present, int *finished)
Decode one audio frame from the input file.
#define AVFMT_GLOBALHEADER
Format wants global header.
Definition: avformat.h:483
struct SwrContext * swr_alloc_set_opts(struct SwrContext *s, int64_t out_ch_layout, enum AVSampleFormat out_sample_fmt, int out_sample_rate, int64_t in_ch_layout, enum AVSampleFormat in_sample_fmt, int in_sample_rate, int log_offset, void *log_ctx)
Allocate SwrContext if needed and set/reset common parameters.
Definition: swresample.c:59
AVOutputFormat * av_guess_format(const char *short_name, const char *filename, const char *mime_type)
Return the output format in the list of registered output formats which best matches the provided par...
Definition: format.c:98
static int add_samples_to_fifo(AVAudioFifo *fifo, uint8_t **converted_input_samples, const int frame_size)
Add converted input audio samples to the FIFO buffer for later processing.
static int output_frame(H264Context *h, AVFrame *dst, H264Picture *srcp)
Definition: h264.c:1175
#define AVERROR_EXIT
Immediate exit was requested; the called function should not be restarted.
Definition: error.h:56
Stream structure.
Definition: avformat.h:876
int frame_size
Number of samples per channel in an audio frame.
Definition: avcodec.h:2430
int av_samples_alloc(uint8_t **audio_data, int *linesize, int nb_channels, int nb_samples, enum AVSampleFormat sample_fmt, int align)
Allocate a samples buffer for nb_samples samples, and fill data pointers and linesize accordingly...
Definition: samplefmt.c:171
int frame_size
Definition: mxfenc.c:1821
static int init_fifo(AVAudioFifo **fifo, AVCodecContext *output_codec_context)
Initialize a FIFO buffer for the audio samples to be encoded.
Libavcodec external API header.
void avcodec_free_context(AVCodecContext **avctx)
Free the codec context and everything associated with it and write NULL to the provided pointer...
Definition: options.c:171
#define OUTPUT_BIT_RATE
The output bit rate in kbit/s.
Definition: transcode_aac.c:44
int sample_rate
samples per second
Definition: avcodec.h:2410
main external API structure.
Definition: avcodec.h:1649
AVCodec * avcodec_find_decoder(enum AVCodecID id)
Find a registered decoder with a matching codec ID.
Definition: utils.c:3063
void av_packet_unref(AVPacket *pkt)
Wipe the packet.
Definition: avpacket.c:563
av_cold void swr_free(SwrContext **ss)
Free the given SwrContext and set the pointer to NULL.
Definition: swresample.c:140
int avcodec_parameters_from_context(AVCodecParameters *par, const AVCodecContext *codec)
Fill the parameters struct based on the values from the supplied codec context.
Definition: utils.c:4077
int avcodec_open2(AVCodecContext *avctx, const AVCodec *codec, AVDictionary **options)
Initialize the AVCodecContext to use the given AVCodec.
Definition: utils.c:1208
void avformat_free_context(AVFormatContext *s)
Free an AVFormatContext and all its streams.
Definition: utils.c:4001
static const char * get_error_text(const int error)
Convert an error code into a text message.
Definition: transcode_aac.c:53
int attribute_align_arg swr_convert(struct SwrContext *s, uint8_t *out_arg[SWR_CH_MAX], int out_count, const uint8_t *in_arg[SWR_CH_MAX], int in_count)
Definition: swresample.c:695
int av_read_frame(AVFormatContext *s, AVPacket *pkt)
Return the next frame of a stream.
Definition: utils.c:1631
int av_frame_get_buffer(AVFrame *frame, int align)
Allocate new buffer(s) for audio or video data.
Definition: frame.c:270
static int64_t pts
Global timestamp for the audio frames.
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:198
int av_audio_fifo_write(AVAudioFifo *af, void **data, int nb_samples)
Write data to an AVAudioFifo.
Definition: audio_fifo.c:113
#define OUTPUT_CHANNELS
The number of output channels.
Definition: transcode_aac.c:46
#define AV_CODEC_FLAG_GLOBAL_HEADER
Place global headers in extradata instead of every keyframe.
Definition: avcodec.h:879
int av_strerror(int errnum, char *errbuf, size_t errbuf_size)
Put a description of the AVERROR code errnum in errbuf.
Definition: error.c:105
Main libavformat public API header.
int avformat_find_stream_info(AVFormatContext *ic, AVDictionary **options)
Read packets of a media file to get stream information.
Definition: utils.c:3268
static int write_output_file_trailer(AVFormatContext *output_format_context)
Write the trailer of the output file container.
void av_init_packet(AVPacket *pkt)
Initialize optional fields of a packet with default values.
Definition: avpacket.c:33
int den
denominator
Definition: rational.h:45
void avformat_close_input(AVFormatContext **s)
Close an opened input AVFormatContext.
Definition: utils.c:4037
Audio FIFO Buffer.
int channels
number of audio channels
Definition: avcodec.h:2411
int avformat_open_input(AVFormatContext **ps, const char *url, AVInputFormat *fmt, AVDictionary **options)
Open an input stream and read the header.
Definition: utils.c:471
int64_t av_get_default_channel_layout(int nb_channels)
Return default channel layout for a given number of channels.
int av_write_trailer(AVFormatContext *s)
Write the stream trailer to an output media file and free the file private data.
Definition: mux.c:1184
#define av_freep(p)
static int init_converted_samples(uint8_t ***converted_input_samples, AVCodecContext *output_codec_context, int frame_size)
Initialize a temporary storage for the specified number of audio samples.
AVCodecParameters * codecpar
Definition: avformat.h:1006
enum AVSampleFormat * sample_fmts
array of supported sample formats, or NULL if unknown, array is terminated by -1
Definition: avcodec.h:3565
static int init_output_frame(AVFrame **frame, AVCodecContext *output_codec_context, int frame_size)
Initialize one input frame for writing to the output file.
AVRational time_base
This is the fundamental unit of time (in seconds) in terms of which frame timestamps are represented...
Definition: avformat.h:913
uint8_t ** extended_data
pointers to the data planes/channels.
Definition: frame.h:231
This structure stores compressed data.
Definition: avcodec.h:1557
void av_register_all(void)
Initialize libavformat and register all the muxers, demuxers and protocols.
Definition: allformats.c:44
int avio_closep(AVIOContext **s)
Close the resource accessed by the AVIOContext *s, free it and set the pointer pointing to it to NULL...
Definition: aviobuf.c:1086
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:241
int strict_std_compliance
strictly follow the standard (MPEG-4, ...).
Definition: avcodec.h:2866
av_cold int swr_init(struct SwrContext *s)
Initialize context after user parameters have been set.
Definition: swresample.c:155
static av_cold void cleanup(FlashSV2Context *s)
Definition: flashsv2enc.c:127