Post AFNCqlwwkXfj3yoHbs by jrc03c@merveilles.town
 (DIR) More posts by jrc03c@merveilles.town
 (DIR) Post #AFNCqkJuoArA0dXb8K by jrc03c@merveilles.town
       2022-01-06T00:57:01Z
       
       0 likes, 0 repeats
       
       Merveilles, I need some help. I've got an audio file, and I'd like to make an EQ visualizer that's rendered out to disk as a video. Currently, I'm using Processing. I read in the audio file, run FFT on each video frame's worth of amplitude samples, draw a simple bar graph as a visualization, and save the image to disk. Then, when it's all done, I use ffmpeg to join the audio and the images into a video. Now, that all works, but it's VERY slow. Do you know of anything better suited to this task?
       
 (DIR) Post #AFNCqkmd5PpDRgyWtE by publius@mastodon.sdf.org
       2022-01-12T19:06:35Z
       
       0 likes, 0 repeats
       
       @jrc03c I'm pretty sure you can do this entirely in ffmpeg. At least, there are some audio visualization modules listed (man ffmpeg-plugins), & it makes sense that you could send that to a file instead of to the screen, although I've not tried.
       
 (DIR) Post #AFNCqlwwkXfj3yoHbs by jrc03c@merveilles.town
       2022-01-06T00:58:53Z
       
       0 likes, 0 repeats
       
       (Just to clarify, I actually run FFT across all amplitude chunks *before* starting the drawing process so that I can just have all of the FFT results in memory. But I'm not sure that this is any faster than running FFT immediately before drawing on each frame since the primary bottleneck seems to be writing the image to disk.)
       
 (DIR) Post #AFNOk1bpKSrP3KTqpU by jrc03c@merveilles.town
       2022-01-12T21:19:49Z
       
       0 likes, 0 repeats
       
       @publius Hm, okay, I'll check that out. Thanks!