> The filter will not process the audio until you call mlt_frame_get_audio()
Thank you for the answer. I tried to follow the sources of mlt and shotcut, but without success... I will try your suggestion. >Why not listen to the "consumer-frame-render" event from the consumer and request the level of the frame after all the filters have been processed? Because I want the maximum speed. Actually, I want to draw the peak audio level of an entire clip, like shotcut does. So I inspected shotcut, didn't fully understand it, and thought anyway that I could use the audiolevel filter. This is just the first job, probably I will in future try to implement other batch jobs, for example a scene change detector or other statistics analisys. ------ >Now I have another problem: I would like to process frames >to inspect the audio levels. I thought I can use the audiolevel filter >and ask it to process every single frame. I make a loop for every frame >I am interested in, call >mlt_service_get_frame(...), then call >mlt_filter_process(audiolevel, frame). > >It works, but only if I make a call to sleep(). >Is there a way I can >get frame by frame, and run on them only the filters I >want, without >using a consumer? > >Thank you, >linuxfan > > > > >_______________________________________________ >Mlt-devel mailing list >Mlt-devel@lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/mlt-devel> _______________________________________________ Mlt-devel mailing list Mlt-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mlt-devel
