FFmpeg
af_dialoguenhance.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2022 Paul B Mahol
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 License
8  * 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
14  * GNU Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public License
17  * along with FFmpeg; if not, write to the Free Software Foundation, Inc.,
18  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19  */
20 
22 #include "libavutil/opt.h"
23 #include "libavutil/tx.h"
24 #include "audio.h"
25 #include "avfilter.h"
26 #include "filters.h"
27 #include "formats.h"
28 #include "internal.h"
29 
30 #include <float.h>
31 
33  const AVClass *class;
34 
36 
37  int fft_size;
38  int overlap;
39 
40  void *window;
41  float *window_float;
42  double *window_double;
45 
53 
55 
58 } AudioDialogueEnhanceContext;
59 
60 #define OFFSET(x) offsetof(AudioDialogueEnhanceContext, x)
61 #define FLAGS AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_FILTERING_PARAM | AV_OPT_FLAG_RUNTIME_PARAM
62 
63 static const AVOption dialoguenhance_options[] = {
64  { "original", "set original center factor", OFFSET(original), AV_OPT_TYPE_DOUBLE, {.dbl=1}, 0, 1, FLAGS },
65  { "enhance", "set dialogue enhance factor",OFFSET(enhance), AV_OPT_TYPE_DOUBLE, {.dbl=1}, 0, 3, FLAGS },
66  { "voice", "set voice detection factor", OFFSET(voice), AV_OPT_TYPE_DOUBLE, {.dbl=2}, 2,32, FLAGS },
67  {NULL}
68 };
69 
70 AVFILTER_DEFINE_CLASS(dialoguenhance);
71 
73 {
75  AVFilterChannelLayouts *in_layout = NULL, *out_layout = NULL;
76  int ret;
77 
78  if ((ret = ff_add_format (&formats, AV_SAMPLE_FMT_FLTP )) < 0 ||
80  (ret = ff_set_common_formats (ctx , formats )) < 0 ||
82  (ret = ff_channel_layouts_ref(in_layout, &ctx->inputs[0]->outcfg.channel_layouts)) < 0 ||
84  (ret = ff_channel_layouts_ref(out_layout, &ctx->outputs[0]->incfg.channel_layouts)) < 0)
85  return ret;
86 
88 }
89 
90 #define DEPTH 32
92 
93 #undef DEPTH
94 #define DEPTH 64
96 
98 {
99  AVFilterContext *ctx = inlink->dst;
100  AudioDialogueEnhanceContext *s = ctx->priv;
101  int ret;
102 
103  s->fft_size = inlink->sample_rate > 100000 ? 8192 : inlink->sample_rate > 50000 ? 4096 : 2048;
104  s->overlap = s->fft_size / 4;
105 
106  s->in_frame = ff_get_audio_buffer(inlink, (s->fft_size + 2) * 2);
107  s->center_frame = ff_get_audio_buffer(inlink, (s->fft_size + 2) * 2);
108  s->out_dist_frame = ff_get_audio_buffer(inlink, (s->fft_size + 2) * 2);
109  s->windowed_frame = ff_get_audio_buffer(inlink, (s->fft_size + 2) * 2);
110  s->windowed_out = ff_get_audio_buffer(inlink, (s->fft_size + 2) * 2);
111  s->windowed_prev = ff_get_audio_buffer(inlink, (s->fft_size + 2) * 2);
112  if (!s->in_frame || !s->windowed_out || !s->windowed_prev ||
113  !s->out_dist_frame || !s->windowed_frame || !s->center_frame)
114  return AVERROR(ENOMEM);
115 
116  switch (inlink->format) {
117  case AV_SAMPLE_FMT_FLTP:
118  s->de_stereo = de_stereo_float;
119  ret = de_tx_init_float(ctx);
120  break;
121  case AV_SAMPLE_FMT_DBLP:
122  s->de_stereo = de_stereo_double;
123  ret = de_tx_init_double(ctx);
124  break;
125  }
126 
127  return ret;
128 }
129 
131 {
132  AVFilterContext *ctx = inlink->dst;
133  AVFilterLink *outlink = ctx->outputs[0];
134  AudioDialogueEnhanceContext *s = ctx->priv;
135  AVFrame *out;
136  int ret;
137 
138  out = ff_get_audio_buffer(outlink, s->overlap);
139  if (!out) {
140  ret = AVERROR(ENOMEM);
141  goto fail;
142  }
143 
144  s->in = in;
145  s->de_stereo(ctx, out);
146 
148  out->nb_samples = in->nb_samples;
149  ret = ff_filter_frame(outlink, out);
150 fail:
151  av_frame_free(&in);
152  s->in = NULL;
153  return ret < 0 ? ret : 0;
154 }
155 
157 {
158  AVFilterLink *inlink = ctx->inputs[0];
159  AVFilterLink *outlink = ctx->outputs[0];
160  AudioDialogueEnhanceContext *s = ctx->priv;
161  AVFrame *in = NULL;
162  int ret = 0, status;
163  int64_t pts;
164 
166 
167  ret = ff_inlink_consume_samples(inlink, s->overlap, s->overlap, &in);
168  if (ret < 0)
169  return ret;
170 
171  if (ret > 0) {
172  return filter_frame(inlink, in);
173  } else if (ff_inlink_acknowledge_status(inlink, &status, &pts)) {
174  ff_outlink_set_status(outlink, status, pts);
175  return 0;
176  } else {
177  if (ff_inlink_queued_samples(inlink) >= s->overlap) {
179  } else if (ff_outlink_frame_wanted(outlink)) {
181  }
182  return 0;
183  }
184 }
185 
187 {
188  AudioDialogueEnhanceContext *s = ctx->priv;
189 
190  av_freep(&s->window);
191 
192  av_frame_free(&s->in_frame);
193  av_frame_free(&s->center_frame);
194  av_frame_free(&s->out_dist_frame);
195  av_frame_free(&s->windowed_frame);
196  av_frame_free(&s->windowed_out);
197  av_frame_free(&s->windowed_prev);
198 
199  av_tx_uninit(&s->tx_ctx[0]);
200  av_tx_uninit(&s->tx_ctx[1]);
201  av_tx_uninit(&s->itx_ctx);
202 }
203 
204 static const AVFilterPad inputs[] = {
205  {
206  .name = "default",
207  .type = AVMEDIA_TYPE_AUDIO,
208  .config_props = config_input,
209  },
210 };
211 
213  .name = "dialoguenhance",
214  .description = NULL_IF_CONFIG_SMALL("Audio Dialogue Enhancement."),
215  .priv_size = sizeof(AudioDialogueEnhanceContext),
216  .priv_class = &dialoguenhance_class,
217  .uninit = uninit,
222  .activate = activate,
223  .process_command = ff_filter_process_command,
224 };
formats
formats
Definition: signature.h:48
ff_get_audio_buffer
AVFrame * ff_get_audio_buffer(AVFilterLink *link, int nb_samples)
Request an audio samples buffer with a specific set of permissions.
Definition: audio.c:97
AV_SAMPLE_FMT_FLTP
@ AV_SAMPLE_FMT_FLTP
float, planar
Definition: samplefmt.h:66
AVFilterChannelLayouts
A list of supported channel layouts.
Definition: formats.h:85
AVERROR
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
opt.h
out
FILE * out
Definition: movenc.c:54
AV_CHANNEL_LAYOUT_STEREO
#define AV_CHANNEL_LAYOUT_STEREO
Definition: channel_layout.h:379
inputs
static const AVFilterPad inputs[]
Definition: af_dialoguenhance.c:204
AudioDialogueEnhancementContext::in_frame
AVFrame * in_frame
Definition: af_dialoguenhance.c:47
ff_filter_frame
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1018
ff_channel_layouts_ref
int ff_channel_layouts_ref(AVFilterChannelLayouts *f, AVFilterChannelLayouts **ref)
Add *ref as a new reference to f.
Definition: formats.c:673
AVTXContext
Definition: tx_priv.h:235
inlink
The exact code depends on how similar the blocks are and how related they are to the and needs to apply these operations to the correct inlink or outlink if there are several Macros are available to factor that when no extra processing is inlink
Definition: filter_design.txt:212
av_frame_free
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:160
AVFrame
This structure describes decoded (raw) audio or video data.
Definition: frame.h:375
AudioDialogueEnhancementContext::de_stereo
int(* de_stereo)(AVFilterContext *ctx, AVFrame *out)
Definition: af_dialoguenhance.c:54
AVOption
AVOption.
Definition: opt.h:346
FILTER_QUERY_FUNC
#define FILTER_QUERY_FUNC(func)
Definition: internal.h:159
filter_frame
static int filter_frame(AVFilterLink *inlink, AVFrame *in)
Definition: af_dialoguenhance.c:130
ff_set_common_all_samplerates
int ff_set_common_all_samplerates(AVFilterContext *ctx)
Equivalent to ff_set_common_samplerates(ctx, ff_all_samplerates())
Definition: formats.c:821
float.h
AVFilter::name
const char * name
Filter name.
Definition: avfilter.h:170
AudioDialogueEnhancementContext::overlap
int overlap
Definition: af_dialoguenhance.c:38
FF_FILTER_FORWARD_STATUS_BACK
#define FF_FILTER_FORWARD_STATUS_BACK(outlink, inlink)
Forward the status on an output link to an input link.
Definition: filters.h:199
uninit
static av_cold void uninit(AVFilterContext *ctx)
Definition: af_dialoguenhance.c:186
AVFilterFormats
A list of supported formats for one end of a filter link.
Definition: formats.h:64
formats.h
AudioDialogueEnhancementContext::windowed_prev
AVFrame * windowed_prev
Definition: af_dialoguenhance.c:51
AudioDialogueEnhancementContext::voice
double voice
Definition: af_dialoguenhance.c:35
fail
#define fail()
Definition: checkasm.h:179
AudioDialogueEnhancementContext::prev_vad_float
float prev_vad_float
Definition: af_dialoguenhance.c:43
ff_af_dialoguenhance
const AVFilter ff_af_dialoguenhance
Definition: af_dialoguenhance.c:212
AudioDialogueEnhancementContext::prev_vad_double
double prev_vad_double
Definition: af_dialoguenhance.c:44
pts
static int64_t pts
Definition: transcode_aac.c:643
AudioDialogueEnhancementContext::windowed_out
AVFrame * windowed_out
Definition: af_dialoguenhance.c:50
AV_CHANNEL_LAYOUT_SURROUND
#define AV_CHANNEL_LAYOUT_SURROUND
Definition: channel_layout.h:382
AVFilterPad
A filter pad used for either input or output.
Definition: internal.h:33
AudioDialogueEnhancementContext::tx_ctx
AVTXContext * tx_ctx[2]
Definition: af_dialoguenhance.c:56
AudioDialogueEnhancementContext::windowed_frame
AVFrame * windowed_frame
Definition: af_dialoguenhance.c:49
AudioDialogueEnhancementContext::itx_fn
av_tx_fn itx_fn
Definition: af_dialoguenhance.c:57
av_cold
#define av_cold
Definition: attributes.h:90
ff_set_common_formats
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:867
av_tx_fn
void(* av_tx_fn)(AVTXContext *s, void *out, void *in, ptrdiff_t stride)
Function pointer to a function to perform the transform.
Definition: tx.h:151
ff_outlink_set_status
static void ff_outlink_set_status(AVFilterLink *link, int status, int64_t pts)
Set the status field of a link from the source filter.
Definition: filters.h:189
ff_inlink_request_frame
void ff_inlink_request_frame(AVFilterLink *link)
Mark that a frame is wanted on the link.
Definition: avfilter.c:1571
s
#define s(width, name)
Definition: cbs_vp9.c:198
AV_OPT_TYPE_DOUBLE
@ AV_OPT_TYPE_DOUBLE
Definition: opt.h:237
AVMEDIA_TYPE_AUDIO
@ AVMEDIA_TYPE_AUDIO
Definition: avutil.h:202
filters.h
AudioDialogueEnhancementContext::out_dist_frame
AVFrame * out_dist_frame
Definition: af_dialoguenhance.c:48
ctx
AVFormatContext * ctx
Definition: movenc.c:48
FLAGS
#define FLAGS
Definition: af_dialoguenhance.c:61
FILTER_INPUTS
#define FILTER_INPUTS(array)
Definition: internal.h:182
AudioDialogueEnhancementContext::original
double original
Definition: af_dialoguenhance.c:35
AVClass
Describe the class of an AVClass context structure.
Definition: log.h:66
ff_inlink_consume_samples
int ff_inlink_consume_samples(AVFilterLink *link, unsigned min, unsigned max, AVFrame **rframe)
Take samples from the link's FIFO and update the link's stats.
Definition: avfilter.c:1465
NULL
#define NULL
Definition: coverity.c:32
av_frame_copy_props
int av_frame_copy_props(AVFrame *dst, const AVFrame *src)
Copy only "metadata" fields from src to dst.
Definition: frame.c:709
OFFSET
#define OFFSET(x)
Definition: af_dialoguenhance.c:60
ff_add_format
int ff_add_format(AVFilterFormats **avff, int64_t fmt)
Add fmt to the list of media formats contained in *avff.
Definition: formats.c:504
ff_audio_default_filterpad
const AVFilterPad ff_audio_default_filterpad[1]
An AVFilterPad array whose only entry has name "default" and is of type AVMEDIA_TYPE_AUDIO.
Definition: audio.c:33
ff_add_channel_layout
int ff_add_channel_layout(AVFilterChannelLayouts **l, const AVChannelLayout *channel_layout)
Definition: formats.c:521
ff_inlink_acknowledge_status
int ff_inlink_acknowledge_status(AVFilterLink *link, int *rstatus, int64_t *rpts)
Test and acknowledge the change of status on the link.
Definition: avfilter.c:1392
AudioDialogueEnhancementContext::window
void * window
Definition: af_dialoguenhance.c:40
NULL_IF_CONFIG_SMALL
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
Definition: internal.h:106
AudioDialogueEnhancementContext::fft_size
int fft_size
Definition: af_dialoguenhance.c:37
AVChannelLayout
An AVChannelLayout holds information about the channel layout of audio data.
Definition: channel_layout.h:303
dialoguenhance_template.c
ff_filter_process_command
int ff_filter_process_command(AVFilterContext *ctx, const char *cmd, const char *arg, char *res, int res_len, int flags)
Generic processing of user supplied commands that are set in the same way as the filter options.
Definition: avfilter.c:890
AudioDialogueEnhancementContext::window_float
float * window_float
Definition: af_dialoguenhance.c:41
av_tx_uninit
av_cold void av_tx_uninit(AVTXContext **ctx)
Frees a context and sets *ctx to NULL, does nothing when *ctx == NULL.
Definition: tx.c:294
AudioDialogueEnhancementContext
Definition: af_dialoguenhance.c:32
internal.h
AVFrame::nb_samples
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:455
dialoguenhance_options
static const AVOption dialoguenhance_options[]
Definition: af_dialoguenhance.c:63
AVFilterPad::name
const char * name
Pad name.
Definition: internal.h:39
ff_inlink_queued_samples
int ff_inlink_queued_samples(AVFilterLink *link)
Definition: avfilter.c:1420
AudioDialogueEnhancementContext::in
AVFrame * in
Definition: af_dialoguenhance.c:46
AudioDialogueEnhancementContext::tx_fn
av_tx_fn tx_fn
Definition: af_dialoguenhance.c:57
AVFilter
Filter definition.
Definition: avfilter.h:166
ret
ret
Definition: filter_design.txt:187
status
ov_status_e status
Definition: dnn_backend_openvino.c:120
config_input
static int config_input(AVFilterLink *inlink)
Definition: af_dialoguenhance.c:97
channel_layout.h
avfilter.h
AV_SAMPLE_FMT_DBLP
@ AV_SAMPLE_FMT_DBLP
double, planar
Definition: samplefmt.h:67
query_formats
static int query_formats(AVFilterContext *ctx)
Definition: af_dialoguenhance.c:72
AVFilterContext
An instance of a filter.
Definition: avfilter.h:407
activate
static int activate(AVFilterContext *ctx)
Definition: af_dialoguenhance.c:156
AudioDialogueEnhancementContext::window_double
double * window_double
Definition: af_dialoguenhance.c:42
audio.h
AudioDialogueEnhancementContext::itx_ctx
AVTXContext * itx_ctx
Definition: af_dialoguenhance.c:56
FILTER_OUTPUTS
#define FILTER_OUTPUTS(array)
Definition: internal.h:183
av_freep
#define av_freep(p)
Definition: tableprint_vlc.h:34
AVFILTER_FLAG_SUPPORT_TIMELINE_INTERNAL
#define AVFILTER_FLAG_SUPPORT_TIMELINE_INTERNAL
Same as AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC, except that the filter will have its filter_frame() c...
Definition: avfilter.h:155
ff_outlink_frame_wanted
the definition of that something depends on the semantic of the filter The callback must examine the status of the filter s links and proceed accordingly The status of output links is stored in the status_in and status_out fields and tested by the ff_outlink_frame_wanted() function. If this function returns true
AudioDialogueEnhancementContext::center_frame
AVFrame * center_frame
Definition: af_dialoguenhance.c:52
int
int
Definition: ffmpeg_filter.c:409
AVFILTER_DEFINE_CLASS
AVFILTER_DEFINE_CLASS(dialoguenhance)
AudioDialogueEnhancementContext::enhance
double enhance
Definition: af_dialoguenhance.c:35
ff_filter_set_ready
void ff_filter_set_ready(AVFilterContext *filter, unsigned priority)
Mark a filter ready and schedule it for activation.
Definition: avfilter.c:234
tx.h