FFmpeg
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
af_earwax.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2011 Mina Nagy Zaki
3  * Copyright (c) 2000 Edward Beingessner And Sundry Contributors.
4  * This source code is freely redistributable and may be used for any purpose.
5  * This copyright notice must be maintained. Edward Beingessner And Sundry
6  * Contributors are not responsible for the consequences of using this
7  * software.
8  *
9  * This file is part of FFmpeg.
10  *
11  * FFmpeg is free software; you can redistribute it and/or
12  * modify it under the terms of the GNU Lesser General Public
13  * License as published by the Free Software Foundation; either
14  * version 2.1 of the License, or (at your option) any later version.
15  *
16  * FFmpeg is distributed in the hope that it will be useful,
17  * but WITHOUT ANY WARRANTY; without even the implied warranty of
18  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19  * Lesser General Public License for more details.
20  *
21  * You should have received a copy of the GNU Lesser General Public
22  * License along with FFmpeg; if not, write to the Free Software
23  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
24  */
25 
26 /**
27  * @file
28  * Stereo Widening Effect. Adds audio cues to move stereo image in
29  * front of the listener. Adapted from the libsox earwax effect.
30  */
31 
33 #include "avfilter.h"
34 #include "audio.h"
35 #include "formats.h"
36 
37 #define NUMTAPS 64
38 
39 static const int8_t filt[NUMTAPS] = {
40 /* 30° 330° */
41  4, -6, /* 32 tap stereo FIR filter. */
42  4, -11, /* One side filters as if the */
43  -1, -5, /* signal was from 30 degrees */
44  3, 3, /* from the ear, the other as */
45  -2, 5, /* if 330 degrees. */
46  -5, 0,
47  9, 1,
48  6, 3, /* Input */
49  -4, -1, /* Left Right */
50  -5, -3, /* __________ __________ */
51  -2, -5, /* | | | | */
52  -7, 1, /* .---| Hh,0(f) | | Hh,0(f) |---. */
53  6, -7, /* / |__________| |__________| \ */
54  30, -29, /* / \ / \ */
55  12, -3, /* / X \ */
56  -11, 4, /* / / \ \ */
57  -3, 7, /* ____V_____ __________V V__________ _____V____ */
58  -20, 23, /* | | | | | | | | */
59  2, 0, /* | Hh,30(f) | | Hh,330(f)| | Hh,330(f)| | Hh,30(f) | */
60  1, -6, /* |__________| |__________| |__________| |__________| */
61  -14, -5, /* \ ___ / \ ___ / */
62  15, -18, /* \ / \ / _____ \ / \ / */
63  6, 7, /* `->| + |<--' / \ `-->| + |<-' */
64  15, -10, /* \___/ _/ \_ \___/ */
65  -14, 22, /* \ / \ / \ / */
66  -7, -2, /* `--->| | | |<---' */
67  -4, 9, /* \_/ \_/ */
68  6, -12, /* */
69  6, -6, /* Headphones */
70  0, -11,
71  0, -5,
72  4, 0};
73 
74 typedef struct {
75  int16_t taps[NUMTAPS * 2];
77 
79 {
80  static const int sample_rates[] = { 44100, -1 };
81 
84 
86  ff_set_common_formats(ctx, formats);
88  ff_set_common_channel_layouts(ctx, layout);
90 
91  return 0;
92 }
93 
94 //FIXME: replace with DSPContext.scalarproduct_int16
95 static inline int16_t *scalarproduct(const int16_t *in, const int16_t *endin, int16_t *out)
96 {
98  int16_t j;
99 
100  while (in < endin) {
101  sample = 0;
102  for (j = 0; j < NUMTAPS; j++)
103  sample += in[j] * filt[j];
104  *out = av_clip_int16(sample >> 6);
105  out++;
106  in++;
107  }
108 
109  return out;
110 }
111 
112 static int filter_frame(AVFilterLink *inlink, AVFrame *insamples)
113 {
114  AVFilterLink *outlink = inlink->dst->outputs[0];
115  int16_t *taps, *endin, *in, *out;
116  AVFrame *outsamples = ff_get_audio_buffer(inlink, insamples->nb_samples);
117  int len;
118 
119  if (!outsamples) {
120  av_frame_free(&insamples);
121  return AVERROR(ENOMEM);
122  }
123  av_frame_copy_props(outsamples, insamples);
124 
125  taps = ((EarwaxContext *)inlink->dst->priv)->taps;
126  out = (int16_t *)outsamples->data[0];
127  in = (int16_t *)insamples ->data[0];
128 
129  len = FFMIN(NUMTAPS, 2*insamples->nb_samples);
130  // copy part of new input and process with saved input
131  memcpy(taps+NUMTAPS, in, len * sizeof(*taps));
132  out = scalarproduct(taps, taps + len, out);
133 
134  // process current input
135  if (2*insamples->nb_samples >= NUMTAPS ){
136  endin = in + insamples->nb_samples * 2 - NUMTAPS;
137  scalarproduct(in, endin, out);
138 
139  // save part of input for next round
140  memcpy(taps, endin, NUMTAPS * sizeof(*taps));
141  } else
142  memmove(taps, taps + 2*insamples->nb_samples, NUMTAPS * sizeof(*taps));
143 
144  av_frame_free(&insamples);
145  return ff_filter_frame(outlink, outsamples);
146 }
147 
148 static const AVFilterPad earwax_inputs[] = {
149  {
150  .name = "default",
151  .type = AVMEDIA_TYPE_AUDIO,
152  .filter_frame = filter_frame,
153  },
154  { NULL }
155 };
156 
157 static const AVFilterPad earwax_outputs[] = {
158  {
159  .name = "default",
160  .type = AVMEDIA_TYPE_AUDIO,
161  },
162  { NULL }
163 };
164 
166  .name = "earwax",
167  .description = NULL_IF_CONFIG_SMALL("Widen the stereo image."),
168  .query_formats = query_formats,
169  .priv_size = sizeof(EarwaxContext),
170  .inputs = earwax_inputs,
171  .outputs = earwax_outputs,
172 };
static const AVFilterPad earwax_outputs[]
Definition: af_earwax.c:157
#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:523
This structure describes decoded (raw) audio or video data.
Definition: frame.h:171
static const AVFilterPad outputs[]
Definition: af_ashowinfo.c:248
Main libavfilter public API header.
static enum AVSampleFormat formats[]
#define AV_CH_LAYOUT_STEREO
#define sample
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:283
const char * name
Pad name.
Definition: internal.h:69
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1158
static const AVFilterPad earwax_inputs[]
Definition: af_earwax.c:148
static int query_formats(AVFilterContext *ctx)
Definition: af_earwax.c:78
A filter pad used for either input or output.
Definition: internal.h:63
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:542
static int16_t * scalarproduct(const int16_t *in, const int16_t *endin, int16_t *out)
Definition: af_earwax.c:95
int ff_add_channel_layout(AVFilterChannelLayouts **l, uint64_t channel_layout)
Definition: formats.c:329
AVFrame * ff_get_audio_buffer(AVFilterLink *link, int nb_samples)
Request an audio samples buffer with a specific set of permissions.
Definition: audio.c:74
#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:148
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:175
void * priv
private data for use by the filter
Definition: avfilter.h:654
int ff_add_format(AVFilterFormats **avff, int64_t fmt)
Add fmt to the list of media formats contained in *avff.
Definition: formats.c:323
static int filter_frame(AVFilterLink *inlink, AVFrame *insamples)
Definition: af_earwax.c:112
static const int sample_rates[]
Definition: dcaenc.h:32
audio channel layout utility functions
#define FFMIN(a, b)
Definition: common.h:81
AVFilter ff_af_earwax
Definition: af_earwax.c:165
int32_t
#define NUMTAPS
Definition: af_earwax.c:37
A list of supported channel layouts.
Definition: formats.h:85
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(constint16_t *) pi >>8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(constint32_t *) pi >>24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(constfloat *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(constfloat *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(constfloat *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(constdouble *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(constdouble *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(constdouble *) pi *(1U<< 31))))#defineSET_CONV_FUNC_GROUP(ofmt, ifmt) staticvoidset_generic_function(AudioConvert *ac){}voidff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enumAVSampleFormatout_fmt, enumAVSampleFormatin_fmt, intchannels, intsample_rate, intapply_map){AudioConvert *ac;intin_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) returnNULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt)>2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);returnNULL;}returnac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}elseif(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;elseac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);returnac;}intff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){intuse_generic=1;intlen=in->nb_samples;intp;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%dsamples-audio_convert:%sto%s(dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));returnff_convert_dither(ac-> in
Filter definition.
Definition: avfilter.h:470
static const AVFilterPad inputs[]
Definition: af_ashowinfo.c:239
const char * name
Filter name.
Definition: avfilter.h:474
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:648
static const int8_t filt[NUMTAPS]
Definition: af_earwax.c:39
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:182
signed 16 bits
Definition: samplefmt.h:62
int len
A list of supported formats for one end of a filter link.
Definition: formats.h:64
uint64_t layout
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(constint16_t *) pi >>8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(constint32_t *) pi >>24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(constfloat *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(constfloat *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(constfloat *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(constdouble *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(constdouble *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(constdouble *) pi *(1U<< 31))))#defineSET_CONV_FUNC_GROUP(ofmt, ifmt) staticvoidset_generic_function(AudioConvert *ac){}voidff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enumAVSampleFormatout_fmt, enumAVSampleFormatin_fmt, intchannels, intsample_rate, intapply_map){AudioConvert *ac;intin_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) returnNULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt)>2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);returnNULL;}returnac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}elseif(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;elseac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);returnac;}intff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){intuse_generic=1;intlen=in->nb_samples;intp;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%dsamples-audio_convert:%sto%s(dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));returnff_convert_dither(ac-> out
An instance of a filter.
Definition: avfilter.h:633
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:225
int ff_set_common_samplerates(AVFilterContext *ctx, AVFilterFormats *samplerates)
Definition: formats.c:530
int av_frame_copy_props(AVFrame *dst, const AVFrame *src)
Copy only "metadata" fields from src to dst.
Definition: frame.c:553