[FFmpeg-devel] [PATCH] lavfi: add aecho filter

Stefano Sabatini stefasab at gmail.com
Tue Jul 9 20:49:14 CEST 2013


On date Tuesday 2013-07-09 14:53:14 +0000, Paul B Mahol encoded:
> Signed-off-by: Paul B Mahol <onemda at gmail.com>
> ---
>  doc/filters.texi         |  29 +++++
>  libavfilter/Makefile     |   1 +
>  libavfilter/af_aecho.c   | 297 +++++++++++++++++++++++++++++++++++++++++++++++
>  libavfilter/allfilters.c |   1 +
>  4 files changed, 328 insertions(+)
>  create mode 100644 libavfilter/af_aecho.c
> 
> diff --git a/doc/filters.texi b/doc/filters.texi
> index 234ff2e..9472675 100644
> --- a/doc/filters.texi
> +++ b/doc/filters.texi
> @@ -347,6 +347,35 @@ aconvert=u8:auto
>  @end example
>  @end itemize
>  
> + at section aecho
> +
> +Apply echoing to the input audio.
> +
> +Echoes are reflected sound and can occur naturally amongst mountains
> +(and sometimes large buildings) when talking or shouting; digital echo
> +effects emulate this behaviour and are often used to help fill out the
> +sound of a single instrument or vocal. The time difference between the
> +original signal and the reflection is the @code{delay}, and the
> +loudness of the reflected signal is the @code{decay}.
> +Multiple echoes can have different delays and decays.
> +
> +A description of the accepted parameters follows.
> +
> + at table @option
> + at item in_gain

> +Set input volume of reflected signal.

Expressed in? Defaul value is?

> +
> + at item out_gain
> +Set output volume of reflected signal.

Same

> +
> + at item delay1..7
> +Set time interval in miliseconds between original signal and reflection.

milliseconds

Also what's "delay1..7"??

> +
> + at item decay1..7
> +Loudness of reflected signal.

Set loudness ...

Same considerations as above.

> +
> + at end table
> +
>  @section afade
>  
>  Apply fade-in/out effect to input audio.
> diff --git a/libavfilter/Makefile b/libavfilter/Makefile
> index cf76ee1..306b24c 100644
> --- a/libavfilter/Makefile
> +++ b/libavfilter/Makefile
> @@ -52,6 +52,7 @@ OBJS-$(CONFIG_AVFORMAT)                      += lavfutils.o
>  OBJS-$(CONFIG_SWSCALE)                       += lswsutils.o
>  
>  OBJS-$(CONFIG_ACONVERT_FILTER)               += af_aconvert.o
> +OBJS-$(CONFIG_AECHO_FILTER)                  += af_aecho.o
>  OBJS-$(CONFIG_AFADE_FILTER)                  += af_afade.o
>  OBJS-$(CONFIG_AFORMAT_FILTER)                += af_aformat.o
>  OBJS-$(CONFIG_AINTERLEAVE_FILTER)            += f_interleave.o
> diff --git a/libavfilter/af_aecho.c b/libavfilter/af_aecho.c
> new file mode 100644
> index 0000000..9d4010d
> --- /dev/null
> +++ b/libavfilter/af_aecho.c
> @@ -0,0 +1,297 @@
> +/*
> + * Copyright (c) 2013 Paul B Mahol
> + *
> + * This file is part of FFmpeg.
> + *
> + * FFmpeg is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU Lesser General Public
> + * License as published by the Free Software Foundation; either
> + * version 2.1 of the License, or (at your option) any later version.
> + *
> + * FFmpeg is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> + * Lesser General Public License for more details.
> + *
> + * You should have received a copy of the GNU Lesser General Public
> + * License along with FFmpeg; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
> + *
> + */
> +
> +#include "libavutil/avstring.h"
> +#include "libavutil/opt.h"
> +#include "libavutil/samplefmt.h"
> +#include "libavutil/avassert.h"
> +#include "avfilter.h"
> +#include "audio.h"
> +#include "internal.h"
> +

> +#define MAX_ECHOS 7

I hate hardcoded arbitrary ranges. Can't you make this value dynamic?
+1 for psychedelic audio/video effects.

> +
> +typedef struct AudioEchoContext {
> +    const AVClass *class;
> +    float in_gain, out_gain;
> +    float delay[MAX_ECHOS], decay[MAX_ECHOS];
> +    int nb_echos;
> +    int delay_index;
> +    uint8_t **delayptrs;
> +    int max_samples, fade_out;
> +    int samples[MAX_ECHOS];
> +    int64_t next_pts;
> +
> +    void (*echo_samples)(struct AudioEchoContext *ctx, uint8_t **delayptrs,
> +                         uint8_t * const *src, uint8_t **dst,
> +                         int nb_samples, int channels);
> +} AudioEchoContext;
> +
> +#define OFFSET(x) offsetof(AudioEchoContext, x)
> +#define A AV_OPT_FLAG_AUDIO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
> +
> +static const AVOption aecho_options[] = {
> +    { "in_gain",  "", OFFSET(in_gain),  AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "out_gain", "", OFFSET(out_gain), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "delay1",   "", OFFSET(delay[0]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 90000, A },
> +    { "decay1",   "", OFFSET(decay[0]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "delay2",   "", OFFSET(delay[1]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 90000, A },
> +    { "decay2",   "", OFFSET(decay[1]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "delay3",   "", OFFSET(delay[2]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 90000, A },
> +    { "decay3",   "", OFFSET(decay[2]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "delay4",   "", OFFSET(delay[3]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 90000, A },
> +    { "decay4",   "", OFFSET(decay[3]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "delay5",   "", OFFSET(delay[4]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 90000, A },
> +    { "decay5",   "", OFFSET(decay[4]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "delay6",   "", OFFSET(delay[5]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 90000, A },
> +    { "decay6",   "", OFFSET(decay[5]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },
> +    { "delay7",   "", OFFSET(delay[6]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 90000, A },
> +    { "decay7",   "", OFFSET(decay[6]), AV_OPT_TYPE_FLOAT, {.dbl=0}, 0, 1, A },

My suggestion is to have a delays=1,2,3,5,8,... decays=1,...

Check example parsing code in lavf/segment.c.

> +    { NULL },
> +};
> +
> +AVFILTER_DEFINE_CLASS(aecho);

nit: aecho is weird, "echo" sounds fine to me

[...]
-- 
FFmpeg = Fierce and Friendly Murdering Proud Ecumenical God


More information about the ffmpeg-devel mailing list