Go to the documentation of this file.
31 #define HISTOGRAM_SIZE 8192
32 #define HISTOGRAM_MAX (HISTOGRAM_SIZE-1)
34 #define MEASURE_ALL UINT_MAX
35 #define MEASURE_NONE 0
37 #define MEASURE_DC_OFFSET (1 << 0)
38 #define MEASURE_MIN_LEVEL (1 << 1)
39 #define MEASURE_MAX_LEVEL (1 << 2)
40 #define MEASURE_MIN_DIFFERENCE (1 << 3)
41 #define MEASURE_MAX_DIFFERENCE (1 << 4)
42 #define MEASURE_MEAN_DIFFERENCE (1 << 5)
43 #define MEASURE_RMS_DIFFERENCE (1 << 6)
44 #define MEASURE_PEAK_LEVEL (1 << 7)
45 #define MEASURE_RMS_LEVEL (1 << 8)
46 #define MEASURE_RMS_PEAK (1 << 9)
47 #define MEASURE_RMS_TROUGH (1 << 10)
48 #define MEASURE_CREST_FACTOR (1 << 11)
49 #define MEASURE_FLAT_FACTOR (1 << 12)
50 #define MEASURE_PEAK_COUNT (1 << 13)
51 #define MEASURE_BIT_DEPTH (1 << 14)
52 #define MEASURE_DYNAMIC_RANGE (1 << 15)
53 #define MEASURE_ZERO_CROSSINGS (1 << 16)
54 #define MEASURE_ZERO_CROSSINGS_RATE (1 << 17)
55 #define MEASURE_NUMBER_OF_SAMPLES (1 << 18)
56 #define MEASURE_NUMBER_OF_NANS (1 << 19)
57 #define MEASURE_NUMBER_OF_INFS (1 << 20)
58 #define MEASURE_NUMBER_OF_DENORMALS (1 << 21)
59 #define MEASURE_NOISE_FLOOR (1 << 22)
60 #define MEASURE_NOISE_FLOOR_COUNT (1 << 23)
61 #define MEASURE_ENTROPY (1 << 24)
62 #define MEASURE_ABS_PEAK_COUNT (1 << 25)
64 #define MEASURE_MINMAXPEAK (MEASURE_MIN_LEVEL | MEASURE_MAX_LEVEL | MEASURE_PEAK_LEVEL)
119 #define OFFSET(x) offsetof(AudioStatsContext, x)
120 #define FLAGS AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
125 {
"reset",
"Set the number of frames over which cumulative stats are calculated before being reset",
OFFSET(reset_count),
AV_OPT_TYPE_INT, {.i64=0}, 0, INT_MAX,
FLAGS },
165 for (
c = 0;
c <
s->nb_channels;
c++) {
204 for (
int n = 0; n <
s->tc_samples; n++)
220 for (
int i = 0;
i <
s->nb_channels;
i++) {
248 unsigned result =
s->maxbitdepth;
249 uint64_t amask =
mask[1] & (~
mask[2]);
253 depth[0] += !!(
mask[0] & (1ULL <<
i));
257 depth[1] += !!(
mask[1] & (1ULL <<
i));
260 for (
int i = 0;
i <
result && !(amask & 1);
i++) {
267 depth[3] += !!(
mask[3] & (1ULL <<
i));
285 int n,
int *ffront,
int *bback)
287 double r, ax =
fabs(x);
290 int empty = front == back &&
ss[front] == -1.0;
292 if (!empty &&
fabs(px) ==
ss[front]) {
299 empty = front == back;
302 if (!empty && ax >=
ss[front]) {
315 while (!empty && ax >=
ss[back]) {
351 double abs_d =
FFABS(d);
352 double drop, noise_floor;
367 }
else if (d == p->
min) {
370 }
else if (p->
last == p->
min) {
383 }
else if (d == p->
max) {
386 }
else if (p->
last == p->
max) {
404 p->
mask[0] |= (
i < 0) ? -
i :
i;
435 if (noise_floor < p->noise_floor) {
447 int type = fpclassify(d);
456 int type = fpclassify(d);
464 const char *fmt,
double val)
471 snprintf(key2,
sizeof(key2),
"lavfi.astats.%d.%s", chan,
key);
473 snprintf(key2,
sizeof(key2),
"lavfi.astats.%s",
key);
477 #define LINEAR_TO_DB(x) (log10(x) * 20)
481 uint64_t
mask[4], min_count = 0, max_count = 0, nb_samples = 0, noise_floor_count = 0;
482 uint64_t nb_nans = 0, nb_infs = 0, nb_denormals = 0;
483 uint64_t abs_peak_count = 0;
484 double min_runs = 0, max_runs = 0,
485 min = DBL_MAX,
max =-DBL_MAX, min_diff = DBL_MAX, max_diff = 0,
486 nmin = DBL_MAX, nmax =-DBL_MAX,
493 min_sigma_x2 = DBL_MAX,
494 max_sigma_x2 =-DBL_MAX;
503 for (
c = 0;
c <
s->nb_channels;
c++) {
578 set_meta(metadata,
c + 1,
"Bit_depth",
"%f", depth[0]);
579 set_meta(metadata,
c + 1,
"Bit_depth2",
"%f", depth[1]);
580 set_meta(metadata,
c + 1,
"Bit_depth3",
"%f", depth[2]);
581 set_meta(metadata,
c + 1,
"Bit_depth4",
"%f", depth[3]);
598 set_meta(metadata, 0,
"Overall.DC_offset",
"%f", max_sigma_x / (nb_samples /
s->nb_channels));
600 set_meta(metadata, 0,
"Overall.Min_level",
"%f",
min);
602 set_meta(metadata, 0,
"Overall.Max_level",
"%f",
max);
604 set_meta(metadata, 0,
"Overall.Min_difference",
"%f", min_diff);
606 set_meta(metadata, 0,
"Overall.Max_difference",
"%f", max_diff);
608 set_meta(metadata, 0,
"Overall.Mean_difference",
"%f", diff1_sum / (nb_samples -
s->nb_channels));
610 set_meta(metadata, 0,
"Overall.RMS_difference",
"%f", sqrt(diff1_sum_x2 / (nb_samples -
s->nb_channels)));
620 set_meta(metadata, 0,
"Overall.Flat_factor",
"%f",
LINEAR_TO_DB((min_runs + max_runs) / (min_count + max_count)));
622 set_meta(metadata, 0,
"Overall.Peak_count",
"%f", (
float)(min_count + max_count) / (
double)
s->nb_channels);
624 set_meta(metadata, 0,
"Overall.Abs_Peak_count",
"%f", (
float)(abs_peak_count) / (
double)
s->nb_channels);
628 set_meta(metadata, 0,
"Overall.Noise_floor_count",
"%f", noise_floor_count / (
double)
s->nb_channels);
630 set_meta(metadata, 0,
"Overall.Entropy",
"%f", entropy / (
double)
s->nb_channels);
633 set_meta(metadata, 0,
"Overall.Bit_depth",
"%f", depth[0]);
634 set_meta(metadata, 0,
"Overall.Bit_depth2",
"%f", depth[1]);
635 set_meta(metadata, 0,
"Overall.Bit_depth3",
"%f", depth[2]);
636 set_meta(metadata, 0,
"Overall.Bit_depth4",
"%f", depth[3]);
639 set_meta(metadata, 0,
"Overall.Number_of_samples",
"%f", nb_samples /
s->nb_channels);
641 set_meta(metadata, 0,
"Number of NaNs",
"%f", nb_nans / (
float)
s->nb_channels);
643 set_meta(metadata, 0,
"Number of Infs",
"%f", nb_infs / (
float)
s->nb_channels);
645 set_meta(metadata, 0,
"Number of denormals",
"%f", nb_denormals / (
float)
s->nb_channels);
648 #define UPDATE_STATS_P(type, update_func, update_float, channel_func) \
649 for (int c = start; c < end; c++) { \
650 ChannelStats *p = &s->chstats[c]; \
651 const type *src = (const type *)data[c]; \
652 const type * const srcend = src + samples; \
653 for (; src < srcend; src++) { \
660 #define UPDATE_STATS_I(type, update_func, update_float, channel_func) \
661 for (int c = start; c < end; c++) { \
662 ChannelStats *p = &s->chstats[c]; \
663 const type *src = (const type *)data[0]; \
664 const type * const srcend = src + samples * channels; \
665 for (src += c; src < srcend; src += channels) { \
672 #define UPDATE_STATS(planar, type, sample, normalizer_suffix, int_sample) \
673 if ((s->measure_overall | s->measure_perchannel) & ~MEASURE_MINMAXPEAK) { \
674 UPDATE_STATS_##planar(type, update_stat(s, p, sample, sample normalizer_suffix, int_sample), s->is_float ? update_float_stat(s, p, sample) : s->is_double ? update_double_stat(s, p, sample) : (void)NULL, ); \
676 UPDATE_STATS_##planar(type, update_minmax(s, p, sample), , p->nmin = p->min normalizer_suffix; p->nmax = p->max normalizer_suffix;); \
684 const uint8_t *
const *
const data = (
const uint8_t *
const *)buf->
extended_data;
732 if (
s->reset_count > 0) {
733 if (
s->nb_frames >=
s->reset_count) {
754 uint64_t
mask[4], min_count = 0, max_count = 0, nb_samples = 0, noise_floor_count = 0;
755 uint64_t nb_nans = 0, nb_infs = 0, nb_denormals = 0, abs_peak_count = 0;
756 double min_runs = 0, max_runs = 0,
757 min = DBL_MAX,
max =-DBL_MAX, min_diff = DBL_MAX, max_diff = 0,
758 nmin = DBL_MAX, nmax =-DBL_MAX,
765 min_sigma_x2 = DBL_MAX,
766 max_sigma_x2 =-DBL_MAX;
775 for (
c = 0;
c <
s->nb_channels;
c++) {
872 if (nb_samples == 0 && !
s->used)
890 av_log(
ctx,
AV_LOG_INFO,
"RMS difference: %f\n", sqrt(diff1_sum_x2 / (nb_samples -
s->nb_channels)));
898 if (min_sigma_x2 != 1)
933 for (
int i = 0;
i <
s->nb_channels;
i++) {
963 .priv_class = &astats_class,
@ AV_SAMPLE_FMT_FLTP
float, planar
static void set_meta(AVDictionary **metadata, int chan, const char *key, const char *fmt, double val)
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
#define MEASURE_PEAK_COUNT
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
static const AVOption astats_options[]
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
#define MEASURE_RMS_TROUGH
#define MEASURE_MIN_LEVEL
#define FILTER_INPUTS(array)
This structure describes decoded (raw) audio or video data.
@ AV_SAMPLE_FMT_S32P
signed 32 bits, planar
static int config_output(AVFilterLink *outlink)
const char * name
Filter name.
int nb_channels
Number of channels in this layout.
static void bit_depth(AudioStatsContext *s, const uint64_t *const mask, uint8_t *depth)
A link between two filters.
static void reset_stats(AudioStatsContext *s)
static void update_stat(AudioStatsContext *s, ChannelStats *p, double d, double nd, int64_t i)
static double calc_noise_floor(double *ss, double x, double px, int n, int *ffront, int *bback)
void * priv
private data for use by the filter
static void update_minmax(AudioStatsContext *s, ChannelStats *p, double d)
static double val(void *priv, double ch)
AVChannelLayout ch_layout
Channel layout of the audio data.
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf type
@ AV_SAMPLE_FMT_S64P
signed 64 bits, planar
#define ss(width, name, subs,...)
uint64_t ehistogram[HISTOGRAM_SIZE]
A filter pad used for either input or output.
#define FILTER_SAMPLEFMTS(...)
#define MEASURE_ABS_PEAK_COUNT
@ AV_OPT_TYPE_DOUBLE
Underlying C type is double.
static int filter_frame(AVFilterLink *inlink, AVFrame *buf)
static double calc_entropy(AudioStatsContext *s, ChannelStats *p)
#define UPDATE_STATS(planar, type, sample, normalizer_suffix, int_sample)
static void update_double_stat(AudioStatsContext *s, ChannelStats *p, double d)
static void update_float_stat(AudioStatsContext *s, ChannelStats *p, float d)
static av_cold void uninit(AVFilterContext *ctx)
#define FILTER_OUTPUTS(array)
#define FFABS(a)
Absolute value, Note, INT_MIN / INT64_MIN result in undefined behavior as they are not representable ...
#define MEASURE_RMS_DIFFERENCE
uint64_t noise_floor_count
Describe the class of an AVClass context structure.
and forward the result(frame or status change) to the corresponding input. If nothing is possible
static __device__ float fabs(float a)
#define MEASURE_NOISE_FLOOR_COUNT
#define MEASURE_ZERO_CROSSINGS_RATE
#define MEASURE_DC_OFFSET
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
#define MEASURE_FLAT_FACTOR
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
#define MEASURE_DYNAMIC_RANGE
int format
agreed upon media format
AVFilterContext * src
source filter
#define MEASURE_NUMBER_OF_SAMPLES
@ AV_SAMPLE_FMT_S16P
signed 16 bits, planar
#define AV_LOG_INFO
Standard information.
static const AVFilterPad astats_inputs[]
int sample_rate
samples per second
#define MEASURE_MEAN_DIFFERENCE
int nb_samples
number of audio samples (per channel) described by this frame
#define i(width, name, range_min, range_max)
#define MEASURE_NUMBER_OF_DENORMALS
int av_get_bytes_per_sample(enum AVSampleFormat sample_fmt)
Return number of bytes per sample.
const AVFilter ff_af_astats
static int filter_channel(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
uint8_t ** extended_data
pointers to the data planes/channels.
int ff_filter_get_nb_threads(AVFilterContext *ctx)
Get number of threads for current filter instance.
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf default value
#define MEASURE_PEAK_LEVEL
@ AV_SAMPLE_FMT_S16
signed 16 bits
static void set_metadata(AudioStatsContext *s, AVDictionary **metadata)
const char * name
Pad name.
void * av_calloc(size_t nmemb, size_t size)
#define MEASURE_RMS_LEVEL
#define MEASURE_NUMBER_OF_NANS
#define MEASURE_BIT_DEPTH
int ff_filter_execute(AVFilterContext *ctx, avfilter_action_func *func, void *arg, int *ret, int nb_jobs)
@ AV_OPT_TYPE_INT
Underlying C type is int.
AVDictionary * metadata
metadata.
#define AVFILTER_FLAG_METADATA_ONLY
The filter is a "metadata" filter - it does not modify the frame data in any way.
@ AV_SAMPLE_FMT_DBLP
double, planar
Filter the word “frame” indicates either a video frame or a group of audio samples
#define MEASURE_NUMBER_OF_INFS
#define AVFILTER_FLAG_SLICE_THREADS
The filter supports multithreading by splitting frames into multiple parts and processing them concur...
#define MEASURE_NOISE_FLOOR
AVChannelLayout ch_layout
channel layout of current buffer (see libavutil/channel_layout.h)
AVFILTER_DEFINE_CLASS(astats)
static void print_stats(AVFilterContext *ctx)
#define MEASURE_MAX_LEVEL
#define MEASURE_MIN_DIFFERENCE
@ AV_OPT_TYPE_BOOL
Underlying C type is int.
int av_dict_set(AVDictionary **pm, const char *key, const char *value, int flags)
Set the given entry in *pm, overwriting an existing entry.
@ AV_OPT_TYPE_FLAGS
Underlying C type is unsigned int.
#define MEASURE_CREST_FACTOR
@ AV_SAMPLE_FMT_DBL
double
@ AV_SAMPLE_FMT_S32
signed 32 bits
@ AV_OPT_TYPE_CONST
Special option type for declaring named constants.
#define MEASURE_MAX_DIFFERENCE
static const AVFilterPad astats_outputs[]
@ AV_SAMPLE_FMT_S64
signed 64 bits
#define MEASURE_ZERO_CROSSINGS