FFmpeg
 All Data Structures Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
af_silencedetect.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2012 Clément Bœsch <ubitux@gmail.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  * Audio silence detector
24  */
25 
26 #include <float.h> /* DBL_MAX */
27 
29 #include "libavutil/opt.h"
30 #include "libavutil/timestamp.h"
31 #include "audio.h"
32 #include "formats.h"
33 #include "avfilter.h"
34 #include "internal.h"
35 
36 typedef struct {
37  const AVClass *class;
38  double noise; ///< noise amplitude ratio
39  double duration; ///< minimum duration of silence until notification
40  int64_t nb_null_samples; ///< current number of continuous zero samples
41  int64_t start; ///< if silence is detected, this value contains the time of the first zero sample
42  int last_sample_rate; ///< last sample rate to check for sample rate changes
44 
45 #define OFFSET(x) offsetof(SilenceDetectContext, x)
46 #define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_AUDIO_PARAM
47 static const AVOption silencedetect_options[] = {
48  { "n", "set noise tolerance", OFFSET(noise), AV_OPT_TYPE_DOUBLE, {.dbl=0.001}, 0, DBL_MAX, FLAGS },
49  { "noise", "set noise tolerance", OFFSET(noise), AV_OPT_TYPE_DOUBLE, {.dbl=0.001}, 0, DBL_MAX, FLAGS },
50  { "d", "set minimum duration in seconds", OFFSET(duration), AV_OPT_TYPE_DOUBLE, {.dbl=2.}, 0, 24*60*60, FLAGS },
51  { "duration", "set minimum duration in seconds", OFFSET(duration), AV_OPT_TYPE_DOUBLE, {.dbl=2.}, 0, 24*60*60, FLAGS },
52  { NULL },
53 };
54 
55 AVFILTER_DEFINE_CLASS(silencedetect);
56 
57 static av_cold int init(AVFilterContext *ctx, const char *args)
58 {
59  int ret;
60  SilenceDetectContext *silence = ctx->priv;
61 
62  silence->class = &silencedetect_class;
63  av_opt_set_defaults(silence);
64 
65  if ((ret = av_set_options_string(silence, args, "=", ":")) < 0)
66  return ret;
67 
68  av_opt_free(silence);
69 
70  return 0;
71 }
72 
73 static char *get_metadata_val(AVFilterBufferRef *insamples, const char *key)
74 {
75  AVDictionaryEntry *e = av_dict_get(insamples->metadata, key, NULL, 0);
76  return e && e->value ? e->value : NULL;
77 }
78 
79 static int filter_frame(AVFilterLink *inlink, AVFilterBufferRef *insamples)
80 {
81  int i;
82  SilenceDetectContext *silence = inlink->dst->priv;
84  const int srate = inlink->sample_rate;
85  const int nb_samples = insamples->audio->nb_samples * nb_channels;
86  const int64_t nb_samples_notify = srate * silence->duration * nb_channels;
87 
88  // scale number of null samples to the new sample rate
89  if (silence->last_sample_rate && silence->last_sample_rate != srate)
90  silence->nb_null_samples =
91  srate * silence->nb_null_samples / silence->last_sample_rate;
92  silence->last_sample_rate = srate;
93 
94  // TODO: support more sample formats
95  // TODO: document metadata
96  if (insamples->format == AV_SAMPLE_FMT_DBL) {
97  double *p = (double *)insamples->data[0];
98 
99  for (i = 0; i < nb_samples; i++, p++) {
100  if (*p < silence->noise && *p > -silence->noise) {
101  if (!silence->start) {
102  silence->nb_null_samples++;
103  if (silence->nb_null_samples >= nb_samples_notify) {
104  silence->start = insamples->pts - (int64_t)(silence->duration / av_q2d(inlink->time_base) + .5);
105  av_dict_set(&insamples->metadata, "lavfi.silence_start",
106  av_ts2timestr(silence->start, &inlink->time_base), 0);
107  av_log(silence, AV_LOG_INFO, "silence_start: %s\n",
108  get_metadata_val(insamples, "lavfi.silence_start"));
109  }
110  }
111  } else {
112  if (silence->start) {
113  av_dict_set(&insamples->metadata, "lavfi.silence_end",
114  av_ts2timestr(insamples->pts, &inlink->time_base), 0);
115  av_dict_set(&insamples->metadata, "lavfi.silence_duration",
116  av_ts2timestr(insamples->pts - silence->start, &inlink->time_base), 0);
117  av_log(silence, AV_LOG_INFO,
118  "silence_end: %s | silence_duration: %s\n",
119  get_metadata_val(insamples, "lavfi.silence_end"),
120  get_metadata_val(insamples, "lavfi.silence_duration"));
121  }
122  silence->nb_null_samples = silence->start = 0;
123  }
124  }
125  }
126 
127  return ff_filter_frame(inlink->dst->outputs[0], insamples);
128 }
129 
131 {
134  static const enum AVSampleFormat sample_fmts[] = {
137  };
138 
139  layouts = ff_all_channel_layouts();
140  if (!layouts)
141  return AVERROR(ENOMEM);
142  ff_set_common_channel_layouts(ctx, layouts);
143 
144  formats = ff_make_format_list(sample_fmts);
145  if (!formats)
146  return AVERROR(ENOMEM);
147  ff_set_common_formats(ctx, formats);
148 
149  formats = ff_all_samplerates();
150  if (!formats)
151  return AVERROR(ENOMEM);
152  ff_set_common_samplerates(ctx, formats);
153 
154  return 0;
155 }
156 
158  {
159  .name = "default",
160  .type = AVMEDIA_TYPE_AUDIO,
161  .get_audio_buffer = ff_null_get_audio_buffer,
162  .filter_frame = filter_frame,
163  },
164  { NULL }
165 };
166 
168  {
169  .name = "default",
170  .type = AVMEDIA_TYPE_AUDIO,
171  },
172  { NULL }
173 };
174 
176  .name = "silencedetect",
177  .description = NULL_IF_CONFIG_SMALL("Detect silence."),
178  .priv_size = sizeof(SilenceDetectContext),
179  .init = init,
181  .inputs = silencedetect_inputs,
182  .outputs = silencedetect_outputs,
183  .priv_class = &silencedetect_class,
184 };