... | ... |
@@ -358,6 +358,36 @@ Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system |
358 | 358 |
that should be preferred (see "-ac" option) unless you have very specific |
359 | 359 |
needs. |
360 | 360 |
|
361 |
+@section silencedetect |
|
362 |
+ |
|
363 |
+Detect silence in an audio stream. |
|
364 |
+ |
|
365 |
+This filter logs a message when it detects that the input audio volume is less |
|
366 |
+or equal to a noise tolerance value for a duration greater or equal to the |
|
367 |
+minimum detected noise duration. |
|
368 |
+ |
|
369 |
+The printed times and duration are expressed in seconds. |
|
370 |
+ |
|
371 |
+@table @option |
|
372 |
+@item duration, d |
|
373 |
+Set silence duration until notification (default is 2 seconds). |
|
374 |
+ |
|
375 |
+@item noise, n |
|
376 |
+Set noise tolerance. Can be specified in dB (in case "dB" is appended to the |
|
377 |
+specified value) or amplitude ratio. Default is -60dB, or 0.001. |
|
378 |
+@end table |
|
379 |
+ |
|
380 |
+Detect 5 seconds of silence with -50dB noise tolerance: |
|
381 |
+@example |
|
382 |
+silencedetect=n=-50dB:d=5 |
|
383 |
+@end example |
|
384 |
+ |
|
385 |
+Complete example with @command{ffmpeg} to detect silence with 0.0001 noise |
|
386 |
+tolerance in @file{silence.mp3}: |
|
387 |
+@example |
|
388 |
+ffmpeg -f lavfi -i amovie=silence.mp3,silencedetect=noise=0.0001 -f null - |
|
389 |
+@end example |
|
390 |
+ |
|
361 | 391 |
@section volume |
362 | 392 |
|
363 | 393 |
Adjust the input audio volume. |
... | ... |
@@ -34,6 +34,7 @@ OBJS-$(CONFIG_ASPLIT_FILTER) += af_asplit.o |
34 | 34 |
OBJS-$(CONFIG_ASTREAMSYNC_FILTER) += af_astreamsync.o |
35 | 35 |
OBJS-$(CONFIG_EARWAX_FILTER) += af_earwax.o |
36 | 36 |
OBJS-$(CONFIG_PAN_FILTER) += af_pan.o |
37 |
+OBJS-$(CONFIG_SILENCEDETECT_FILTER) += af_silencedetect.o |
|
37 | 38 |
OBJS-$(CONFIG_VOLUME_FILTER) += af_volume.o |
38 | 39 |
|
39 | 40 |
OBJS-$(CONFIG_ABUFFER_FILTER) += asrc_abuffer.o |
40 | 41 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,175 @@ |
0 |
+/* |
|
1 |
+ * Copyright (c) 2012 Clément Bœsch <ubitux@gmail.com> |
|
2 |
+ * |
|
3 |
+ * This file is part of FFmpeg. |
|
4 |
+ * |
|
5 |
+ * FFmpeg is free software; you can redistribute it and/or |
|
6 |
+ * modify it under the terms of the GNU Lesser General Public |
|
7 |
+ * License as published by the Free Software Foundation; either |
|
8 |
+ * version 2.1 of the License, or (at your option) any later version. |
|
9 |
+ * |
|
10 |
+ * FFmpeg is distributed in the hope that it will be useful, |
|
11 |
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
12 |
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
|
13 |
+ * Lesser General Public License for more details. |
|
14 |
+ * |
|
15 |
+ * You should have received a copy of the GNU Lesser General Public |
|
16 |
+ * License along with FFmpeg; if not, write to the Free Software |
|
17 |
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
|
18 |
+ */ |
|
19 |
+ |
|
20 |
+/** |
|
21 |
+ * @file |
|
22 |
+ * Audio silence detector |
|
23 |
+ */ |
|
24 |
+ |
|
25 |
+#include "libavutil/opt.h" |
|
26 |
+#include "avfilter.h" |
|
27 |
+ |
|
28 |
+typedef struct { |
|
29 |
+ const AVClass *class; |
|
30 |
+ char *noise_str; ///< noise option string |
|
31 |
+ double noise; ///< noise amplitude ratio |
|
32 |
+ int duration; ///< minimum duration of silence until notification |
|
33 |
+ int64_t nb_null_samples; ///< current number of continuous zero samples |
|
34 |
+ double start; ///< if silence is detected, this value contains the time of the first zero sample |
|
35 |
+ int last_sample_rate; ///< last sample rate to check for sample rate changes |
|
36 |
+} SilenceDetectContext; |
|
37 |
+ |
|
38 |
+#define OFFSET(x) offsetof(SilenceDetectContext, x) |
|
39 |
+static const AVOption silencedetect_options[] = { |
|
40 |
+ { "n", "set noise tolerance", OFFSET(noise_str), AV_OPT_TYPE_STRING, {.str="-60dB"}, CHAR_MIN, CHAR_MAX }, |
|
41 |
+ { "noise", "set noise tolerance", OFFSET(noise_str), AV_OPT_TYPE_STRING, {.str="-60dB"}, CHAR_MIN, CHAR_MAX }, |
|
42 |
+ { "d", "set minimum duration in seconds", OFFSET(duration), AV_OPT_TYPE_INT, {.dbl=2}, 0, INT_MAX}, |
|
43 |
+ { "duration", "set minimum duration in seconds", OFFSET(duration), AV_OPT_TYPE_INT, {.dbl=2}, 0, INT_MAX}, |
|
44 |
+ { NULL }, |
|
45 |
+}; |
|
46 |
+ |
|
47 |
+static const char *silencedetect_get_name(void *ctx) |
|
48 |
+{ |
|
49 |
+ return "silencedetect"; |
|
50 |
+} |
|
51 |
+ |
|
52 |
+static const AVClass silencedetect_class = { |
|
53 |
+ .class_name = "SilenceDetectContext", |
|
54 |
+ .item_name = silencedetect_get_name, |
|
55 |
+ .option = silencedetect_options, |
|
56 |
+}; |
|
57 |
+ |
|
58 |
+static av_cold int init(AVFilterContext *ctx, const char *args, void *opaque) |
|
59 |
+{ |
|
60 |
+ int ret; |
|
61 |
+ char *tail; |
|
62 |
+ SilenceDetectContext *silence = ctx->priv; |
|
63 |
+ |
|
64 |
+ silence->class = &silencedetect_class; |
|
65 |
+ av_opt_set_defaults(silence); |
|
66 |
+ |
|
67 |
+ if ((ret = av_set_options_string(silence, args, "=", ":")) < 0) { |
|
68 |
+ av_log(ctx, AV_LOG_ERROR, "Error parsing options string: '%s'\n", args); |
|
69 |
+ return ret; |
|
70 |
+ } |
|
71 |
+ |
|
72 |
+ silence->noise = strtod(silence->noise_str, &tail); |
|
73 |
+ if (!strcmp(tail, "dB")) { |
|
74 |
+ silence->noise = pow(10, silence->noise/20); |
|
75 |
+ } else if (*tail) { |
|
76 |
+ av_log(ctx, AV_LOG_ERROR, "Invalid value '%s' for noise parameter.\n", |
|
77 |
+ silence->noise_str); |
|
78 |
+ return AVERROR(EINVAL); |
|
79 |
+ } |
|
80 |
+ |
|
81 |
+ return 0; |
|
82 |
+} |
|
83 |
+ |
|
84 |
+static void filter_samples(AVFilterLink *inlink, AVFilterBufferRef *insamples) |
|
85 |
+{ |
|
86 |
+ int i; |
|
87 |
+ SilenceDetectContext *silence = inlink->dst->priv; |
|
88 |
+ const int nb_channels = av_get_channel_layout_nb_channels(inlink->channel_layout); |
|
89 |
+ const int srate = inlink->sample_rate; |
|
90 |
+ const int nb_samples = insamples->audio->nb_samples * nb_channels; |
|
91 |
+ const int64_t nb_samples_notify = srate * silence->duration * nb_channels; |
|
92 |
+ |
|
93 |
+ // scale number of null samples to the new sample rate |
|
94 |
+ if (silence->last_sample_rate && silence->last_sample_rate != srate) |
|
95 |
+ silence->nb_null_samples = |
|
96 |
+ srate * silence->nb_null_samples / silence->last_sample_rate; |
|
97 |
+ silence->last_sample_rate = srate; |
|
98 |
+ |
|
99 |
+ // TODO: support more sample formats |
|
100 |
+ if (insamples->format == AV_SAMPLE_FMT_DBL) { |
|
101 |
+ double *p = (double *)insamples->data[0]; |
|
102 |
+ |
|
103 |
+ for (i = 0; i < nb_samples; i++, p++) { |
|
104 |
+ if (*p < silence->noise && *p > -silence->noise) { |
|
105 |
+ if (!silence->start) { |
|
106 |
+ silence->nb_null_samples++; |
|
107 |
+ if (silence->nb_null_samples >= nb_samples_notify) { |
|
108 |
+ silence->start = insamples->pts * av_q2d(inlink->time_base) - silence->duration; |
|
109 |
+ av_log(silence, AV_LOG_INFO, |
|
110 |
+ "silence_start: %f\n", silence->start); |
|
111 |
+ } |
|
112 |
+ } |
|
113 |
+ } else { |
|
114 |
+ if (silence->start) { |
|
115 |
+ double end = insamples->pts * av_q2d(inlink->time_base); |
|
116 |
+ av_log(silence, AV_LOG_INFO, |
|
117 |
+ "silence_end: %f | silence_duration: %f\n", |
|
118 |
+ end, end - silence->start); |
|
119 |
+ } |
|
120 |
+ silence->nb_null_samples = silence->start = 0; |
|
121 |
+ } |
|
122 |
+ } |
|
123 |
+ } |
|
124 |
+ |
|
125 |
+ avfilter_filter_samples(inlink->dst->outputs[0], insamples); |
|
126 |
+} |
|
127 |
+ |
|
128 |
+static int query_formats(AVFilterContext *ctx) |
|
129 |
+{ |
|
130 |
+ AVFilterFormats *formats = NULL; |
|
131 |
+ enum AVSampleFormat sample_fmts[] = { |
|
132 |
+ AV_SAMPLE_FMT_DBL, |
|
133 |
+ AV_SAMPLE_FMT_NONE |
|
134 |
+ }; |
|
135 |
+ int packing_fmts[] = { AVFILTER_PACKED, -1 }; |
|
136 |
+ |
|
137 |
+ formats = avfilter_make_all_channel_layouts(); |
|
138 |
+ if (!formats) |
|
139 |
+ return AVERROR(ENOMEM); |
|
140 |
+ avfilter_set_common_channel_layouts(ctx, formats); |
|
141 |
+ |
|
142 |
+ formats = avfilter_make_format_list(sample_fmts); |
|
143 |
+ if (!formats) |
|
144 |
+ return AVERROR(ENOMEM); |
|
145 |
+ avfilter_set_common_sample_formats(ctx, formats); |
|
146 |
+ |
|
147 |
+ formats = avfilter_make_format_list(packing_fmts); |
|
148 |
+ if (!formats) |
|
149 |
+ return AVERROR(ENOMEM); |
|
150 |
+ avfilter_set_common_packing_formats(ctx, formats); |
|
151 |
+ |
|
152 |
+ return 0; |
|
153 |
+} |
|
154 |
+ |
|
155 |
+AVFilter avfilter_af_silencedetect = { |
|
156 |
+ .name = "silencedetect", |
|
157 |
+ .description = NULL_IF_CONFIG_SMALL("Detect silence."), |
|
158 |
+ .priv_size = sizeof(SilenceDetectContext), |
|
159 |
+ .init = init, |
|
160 |
+ .query_formats = query_formats, |
|
161 |
+ |
|
162 |
+ .inputs = (const AVFilterPad[]) { |
|
163 |
+ { .name = "default", |
|
164 |
+ .type = AVMEDIA_TYPE_AUDIO, |
|
165 |
+ .get_audio_buffer = avfilter_null_get_audio_buffer, |
|
166 |
+ .filter_samples = filter_samples, }, |
|
167 |
+ { .name = NULL } |
|
168 |
+ }, |
|
169 |
+ .outputs = (const AVFilterPad[]) { |
|
170 |
+ { .name = "default", |
|
171 |
+ .type = AVMEDIA_TYPE_AUDIO, }, |
|
172 |
+ { .name = NULL } |
|
173 |
+ }, |
|
174 |
+}; |
... | ... |
@@ -44,6 +44,7 @@ void avfilter_register_all(void) |
44 | 44 |
REGISTER_FILTER (ASTREAMSYNC, astreamsync, af); |
45 | 45 |
REGISTER_FILTER (EARWAX, earwax, af); |
46 | 46 |
REGISTER_FILTER (PAN, pan, af); |
47 |
+ REGISTER_FILTER (SILENCEDETECT, silencedetect, af); |
|
47 | 48 |
REGISTER_FILTER (VOLUME, volume, af); |
48 | 49 |
|
49 | 50 |
REGISTER_FILTER (ABUFFER, abuffer, asrc); |
... | ... |
@@ -30,8 +30,8 @@ |
30 | 30 |
#include "libavcodec/avcodec.h" |
31 | 31 |
|
32 | 32 |
#define LIBAVFILTER_VERSION_MAJOR 2 |
33 |
-#define LIBAVFILTER_VERSION_MINOR 57 |
|
34 |
-#define LIBAVFILTER_VERSION_MICRO 101 |
|
33 |
+#define LIBAVFILTER_VERSION_MINOR 58 |
|
34 |
+#define LIBAVFILTER_VERSION_MICRO 100 |
|
35 | 35 |
|
36 | 36 |
#define LIBAVFILTER_VERSION_INT AV_VERSION_INT(LIBAVFILTER_VERSION_MAJOR, \ |
37 | 37 |
LIBAVFILTER_VERSION_MINOR, \ |