Hi!
I am trying to build a video editing tool for a local radio station, but also
going to open source it. It's a tool which automatically makes cuts to always
follow (focus) the camera to the person who is speaking. To do this I need to
divide the tracks in X-seconds samples, and calculate which tracks have a
higher volume per sample, and focus that camera in that second. I have a
deadline for the 11th of August.
Here are some example (propietary) applications that do this:
- https://www.kapwing.com/ai/auto-speaker-focus
- https://www.autocut.com/en/autocutpodcast/
- https://www.autopod.fm/
Because I don't know much of the API yet, this is my code in metaprogramming. I
hope the code is understandable. https://pastebin.com/YVxBu95B
Any help appreciated. Thanks!
_______________________________________________
Mlt-devel mailing list
Mlt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlt-devel
