> Doesn't produce similar output as sox. > > Looks like there is no need to use generate_wave_table > as phase is calculated differently, even tremolo from > tap plugins do something different. > > Can you elaborate this approach?
This filter is just amplitude modulation. Just took a listen and tremolo from SoX is the same, but it expects its depth parameter to be a percentage between 0-100. This filter expects the same parameter, but the range is 0.0-1.0. You may be thinking of `vibrato' which is phase modulation, not an amplitude modulation. I could implement a vibrato as a filter as well. I calculate my envelope with ff_generate_wave_table() to avoid having to re-calculate sine for every sample. This is faster, but uses a small amount of memory. On Mon, Sep 21, 2015 at 4:18 PM, Paul B Mahol <[email protected]> wrote: > On 9/21/15, Kyle Swanson <[email protected]> wrote: >> Signed-off-by: Kyle Swanson <[email protected]> >> --- >> doc/filters.texi | 19 +++++ >> libavfilter/Makefile | 1 + >> libavfilter/af_tremolo.c | 177 >> +++++++++++++++++++++++++++++++++++++++++++++++ >> libavfilter/allfilters.c | 1 + >> libavfilter/version.h | 2 +- >> 5 files changed, 199 insertions(+), 1 deletion(-) >> create mode 100644 libavfilter/af_tremolo.c >> > > Doesn't produce similar output as sox. > > Looks like there is no need to use generate_wave_table > as phase is calculated differently, even tremolo from > tap plugins do something different. > > Can you elaborate this approach? > _______________________________________________ > ffmpeg-devel mailing list > [email protected] > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel _______________________________________________ ffmpeg-devel mailing list [email protected] http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
