FFmpeg
af_rubberband.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 #include <rubberband/rubberband-c.h>
20 
22 #include "libavutil/common.h"
23 #include "libavutil/opt.h"
24 
25 #include "audio.h"
26 #include "avfilter.h"
27 #include "filters.h"
28 #include "formats.h"
29 #include "internal.h"
30 
31 typedef struct RubberBandContext {
32  const AVClass *class;
33  RubberBandState rbs;
34 
35  double tempo, pitch;
38  int64_t nb_samples_out;
39  int64_t nb_samples_in;
40  int64_t first_pts;
43 
44 #define OFFSET(x) offsetof(RubberBandContext, x)
45 #define A AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
46 #define AT AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_RUNTIME_PARAM
47 
48 static const AVOption rubberband_options[] = {
49  { "tempo", "set tempo scale factor", OFFSET(tempo), AV_OPT_TYPE_DOUBLE, {.dbl=1}, 0.01, 100, AT },
50  { "pitch", "set pitch scale factor", OFFSET(pitch), AV_OPT_TYPE_DOUBLE, {.dbl=1}, 0.01, 100, AT },
51  { "transients", "set transients", OFFSET(transients), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "transients" },
52  { "crisp", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionTransientsCrisp}, 0, 0, A, "transients" },
53  { "mixed", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionTransientsMixed}, 0, 0, A, "transients" },
54  { "smooth", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionTransientsSmooth}, 0, 0, A, "transients" },
55  { "detector", "set detector", OFFSET(detector), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "detector" },
56  { "compound", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionDetectorCompound}, 0, 0, A, "detector" },
57  { "percussive", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionDetectorPercussive}, 0, 0, A, "detector" },
58  { "soft", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionDetectorSoft}, 0, 0, A, "detector" },
59  { "phase", "set phase", OFFSET(phase), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "phase" },
60  { "laminar", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionPhaseLaminar}, 0, 0, A, "phase" },
61  { "independent", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionPhaseIndependent}, 0, 0, A, "phase" },
62  { "window", "set window", OFFSET(window), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "window" },
63  { "standard", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionWindowStandard}, 0, 0, A, "window" },
64  { "short", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionWindowShort}, 0, 0, A, "window" },
65  { "long", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionWindowLong}, 0, 0, A, "window" },
66  { "smoothing", "set smoothing", OFFSET(smoothing), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "smoothing" },
67  { "off", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionSmoothingOff}, 0, 0, A, "smoothing" },
68  { "on", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionSmoothingOn}, 0, 0, A, "smoothing" },
69  { "formant", "set formant", OFFSET(formant), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "formant" },
70  { "shifted", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionFormantShifted}, 0, 0, A, "formant" },
71  { "preserved", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionFormantPreserved}, 0, 0, A, "formant" },
72  { "pitchq", "set pitch quality", OFFSET(opitch), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "pitch" },
73  { "quality", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionPitchHighQuality}, 0, 0, A, "pitch" },
74  { "speed", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionPitchHighSpeed}, 0, 0, A, "pitch" },
75  { "consistency", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionPitchHighConsistency}, 0, 0, A, "pitch" },
76  { "channels", "set channels", OFFSET(channels), AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX, A, "channels" },
77  { "apart", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionChannelsApart}, 0, 0, A, "channels" },
78  { "together", 0, 0, AV_OPT_TYPE_CONST, {.i64=RubberBandOptionChannelsTogether}, 0, 0, A, "channels" },
79  { NULL },
80 };
81 
82 AVFILTER_DEFINE_CLASS(rubberband);
83 
85 {
86  RubberBandContext *s = ctx->priv;
87 
88  if (s->rbs)
89  rubberband_delete(s->rbs);
90 }
91 
93 {
96  static const enum AVSampleFormat sample_fmts[] = {
99  };
100  int ret;
101 
102  layouts = ff_all_channel_counts();
103  if (!layouts)
104  return AVERROR(ENOMEM);
105  ret = ff_set_common_channel_layouts(ctx, layouts);
106  if (ret < 0)
107  return ret;
108 
109  formats = ff_make_format_list(sample_fmts);
110  if (!formats)
111  return AVERROR(ENOMEM);
112  ret = ff_set_common_formats(ctx, formats);
113  if (ret < 0)
114  return ret;
115 
116  formats = ff_all_samplerates();
117  if (!formats)
118  return AVERROR(ENOMEM);
119  return ff_set_common_samplerates(ctx, formats);
120 }
121 
123 {
124  AVFilterContext *ctx = inlink->dst;
125  RubberBandContext *s = ctx->priv;
126  AVFilterLink *outlink = ctx->outputs[0];
127  AVFrame *out;
128  int ret = 0, nb_samples;
129 
130  if (s->first_pts == AV_NOPTS_VALUE)
131  s->first_pts = in->pts;
132 
133  rubberband_process(s->rbs, (const float *const *)in->data, in->nb_samples, ff_outlink_get_status(inlink));
134  s->nb_samples_in += in->nb_samples;
135 
136  nb_samples = rubberband_available(s->rbs);
137  if (nb_samples > 0) {
138  out = ff_get_audio_buffer(outlink, nb_samples);
139  if (!out) {
140  av_frame_free(&in);
141  return AVERROR(ENOMEM);
142  }
143  out->pts = s->first_pts + av_rescale_q(s->nb_samples_out,
144  (AVRational){ 1, outlink->sample_rate },
145  outlink->time_base);
146  nb_samples = rubberband_retrieve(s->rbs, (float *const *)out->data, nb_samples);
147  out->nb_samples = nb_samples;
148  ret = ff_filter_frame(outlink, out);
150  }
151 
152  av_frame_free(&in);
153  if (ff_inlink_queued_samples(inlink) >= s->nb_samples)
154  ff_filter_set_ready(ctx, 100);
155  return ret < 0 ? ret : nb_samples;
156 }
157 
159 {
160  AVFilterContext *ctx = inlink->dst;
161  RubberBandContext *s = ctx->priv;
162  int opts = s->transients|s->detector|s->phase|s->window|
163  s->smoothing|s->formant|s->opitch|s->channels|
164  RubberBandOptionProcessRealTime;
165 
166  if (s->rbs)
167  rubberband_delete(s->rbs);
168  s->rbs = rubberband_new(inlink->sample_rate, inlink->channels, opts, 1. / s->tempo, s->pitch);
169  if (!s->rbs)
170  return AVERROR(ENOMEM);
171 
172  s->nb_samples = rubberband_get_samples_required(s->rbs);
174 
175  return 0;
176 }
177 
179 {
180  AVFilterLink *inlink = ctx->inputs[0];
181  AVFilterLink *outlink = ctx->outputs[0];
182  RubberBandContext *s = ctx->priv;
183  AVFrame *in = NULL;
184  int ret;
185 
186  FF_FILTER_FORWARD_STATUS_BACK(outlink, inlink);
187 
188  ret = ff_inlink_consume_samples(inlink, s->nb_samples, s->nb_samples, &in);
189  if (ret < 0)
190  return ret;
191  if (ret > 0) {
192  ret = filter_frame(inlink, in);
193  if (ret != 0)
194  return ret;
195  }
196 
197  FF_FILTER_FORWARD_STATUS(inlink, outlink);
198  FF_FILTER_FORWARD_WANTED(outlink, inlink);
199 
200  return FFERROR_NOT_READY;
201 }
202 
203 static int process_command(AVFilterContext *ctx, const char *cmd, const char *args,
204  char *res, int res_len, int flags)
205 {
206  RubberBandContext *s = ctx->priv;
207  int ret;
208 
209  ret = ff_filter_process_command(ctx, cmd, args, res, res_len, flags);
210  if (ret < 0)
211  return ret;
212 
213  rubberband_set_time_ratio(s->rbs, 1. / s->tempo);
214  rubberband_set_pitch_scale(s->rbs, s->pitch);
215 
216  return 0;
217 }
218 
219 static const AVFilterPad rubberband_inputs[] = {
220  {
221  .name = "default",
222  .type = AVMEDIA_TYPE_AUDIO,
223  .config_props = config_input,
224  },
225  { NULL }
226 };
227 
228 static const AVFilterPad rubberband_outputs[] = {
229  {
230  .name = "default",
231  .type = AVMEDIA_TYPE_AUDIO,
232  },
233  { NULL }
234 };
235 
237  .name = "rubberband",
238  .description = NULL_IF_CONFIG_SMALL("Apply time-stretching and pitch-shifting."),
239  .query_formats = query_formats,
240  .priv_size = sizeof(RubberBandContext),
241  .priv_class = &rubberband_class,
242  .uninit = uninit,
243  .activate = activate,
244  .inputs = rubberband_inputs,
245  .outputs = rubberband_outputs,
247 };
static const AVFilterPad rubberband_outputs[]
float, planar
Definition: samplefmt.h:69
int64_t nb_samples_out
Definition: af_rubberband.c:38
#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:550
This structure describes decoded (raw) audio or video data.
Definition: frame.h:295
AVOption.
Definition: opt.h:246
static int config_input(AVFilterLink *inlink)
#define A
Definition: af_rubberband.c:45
Main libavfilter public API header.
channels
Definition: aptx.c:30
FF_FILTER_FORWARD_STATUS(inlink, outlink)
return FFERROR_NOT_READY
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:60
AVFilterLink ** inputs
array of pointers to input links
Definition: avfilter.h:346
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1075
#define av_cold
Definition: attributes.h:82
AVOptions.
static int process_command(AVFilterContext *ctx, const char *cmd, const char *args, char *res, int res_len, int flags)
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:388
#define OFFSET(x)
Definition: af_rubberband.c:44
#define FF_FILTER_FORWARD_STATUS_BACK(outlink, inlink)
Forward the status on an output link to an input link.
Definition: filters.h:199
A filter pad used for either input or output.
Definition: internal.h:54
int64_t av_rescale_q(int64_t a, AVRational bq, AVRational cq)
Rescale a 64-bit integer by 2 rational numbers.
Definition: mathematics.c:142
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:569
static int activate(AVFilterContext *ctx)
AVFrame * ff_get_audio_buffer(AVFilterLink *link, int nb_samples)
Request an audio samples buffer with a specific set of permissions.
Definition: audio.c:86
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:202
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:186
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:869
void * priv
private data for use by the filter
Definition: avfilter.h:353
static const AVOption rubberband_options[]
Definition: af_rubberband.c:48
static const AVFilterPad rubberband_inputs[]
AVDictionary * opts
Definition: movenc.c:50
audio channel layout utility functions
int ff_inlink_queued_samples(AVFilterLink *link)
Definition: avfilter.c:1456
AVFormatContext * ctx
Definition: movenc.c:48
#define s(width, name)
Definition: cbs_vp9.c:257
static int query_formats(AVFilterContext *ctx)
Definition: af_rubberband.c:92
RubberBandState rbs
Definition: af_rubberband.c:33
static const AVFilterPad outputs[]
Definition: af_acontrast.c:203
A list of supported channel layouts.
Definition: formats.h:85
AVSampleFormat
Audio sample formats.
Definition: samplefmt.h:58
int ff_inlink_consume_samples(AVFilterLink *link, unsigned min, unsigned max, AVFrame **rframe)
Take samples from the link&#39;s FIFO and update the link&#39;s stats.
Definition: avfilter.c:1495
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several inputs
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(const int16_t *) pi >> 8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(const int32_t *) pi >> 24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(const float *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(const float *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(const float *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(const double *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(const double *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(const double *) pi *(1U<< 31))))#define SET_CONV_FUNC_GROUP(ofmt, ifmt) static void set_generic_function(AudioConvert *ac){}void ff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enum AVSampleFormat out_fmt, enum AVSampleFormat in_fmt, int channels, int sample_rate, int apply_map){AudioConvert *ac;int in_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) return NULL;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);return NULL;}return ac;}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;}else if(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;else ac->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);return ac;}int ff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){int use_generic=1;int len=in->nb_samples;int p;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%d samples - audio_convert: %s to %s (dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));return ff_convert_dither(ac-> in
Describe the class of an AVClass context structure.
Definition: log.h:67
Filter definition.
Definition: avfilter.h:144
int ff_outlink_get_status(AVFilterLink *link)
Get the status on an output link.
Definition: avfilter.c:1625
Rational number (pair of numerator and denominator).
Definition: rational.h:58
static av_cold void uninit(AVFilterContext *ctx)
Definition: af_rubberband.c:84
const char * name
Filter name.
Definition: avfilter.h:148
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:350
enum MovChannelLayoutTag * layouts
Definition: mov_chan.c:434
AVFilterFormats * ff_all_samplerates(void)
Definition: formats.c:394
#define flags(name, subs,...)
Definition: cbs_av1.c:564
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:309
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
AVFilter ff_af_rubberband
#define AT
Definition: af_rubberband.c:46
common internal and external API header
void ff_filter_set_ready(AVFilterContext *filter, unsigned priority)
Mark a filter ready and schedule it for activation.
Definition: avfilter.c:193
FF_FILTER_FORWARD_WANTED(outlink, inlink)
AVFILTER_DEFINE_CLASS(rubberband)
A list of supported formats for one end of a filter link.
Definition: formats.h:64
An instance of a filter.
Definition: avfilter.h:338
static int filter_frame(AVFilterLink *inlink, AVFrame *in)
static enum AVSampleFormat sample_fmts[]
Definition: adpcmenc.c:701
FILE * out
Definition: movenc.c:54
formats
Definition: signature.h:48
internal API functions
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
AVFilterChannelLayouts * ff_all_channel_counts(void)
Construct an AVFilterChannelLayouts coding for any channel layout, with known or unknown disposition...
Definition: formats.c:409
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:361
int ff_set_common_samplerates(AVFilterContext *ctx, AVFilterFormats *samplerates)
Definition: formats.c:557
#define AV_NOPTS_VALUE
Undefined timestamp value.
Definition: avutil.h:248