[FFmpeg-devel] Speech Recognition Filter Idea

[Harry Muscle]

I'm thinking of developing a speech recognition audio filter for ffmpeg.
The idea would be that this filter would search for specific words or
phrases and then apply another filter on those words.  For example you
could increase the volume whenever a certain word is encountered.  Or you
could mute the volume whenever a swear word is found.  There's probably
lots of other uses but those are the first two that come to mind.  This
filter would require an external library (probably Mozilla Deep Speech) so
I'm not sure if that would prevent it from ever being included in ffmpeg.
Could someone clarify this point?  If the license of the external library
is compatible, would such a filter be considered for inclusion in the
official ffmpeg code (assuming it meets all coding standards, etc.).

Thanks,
Harry