mubu.usual.onseg doesn’t seem to draw all markers in imubu properly until you click in graphic display of imubu
see for example helpfile of mubu.usual.onseg
see also attached image
I am having a similar issue with segmentation with pipo. I automatically read a .wav file to a mubu container. Then, using the read output of the mubu object, I initiate a mubu.process for segmentation, which creates markers of segment locations. However, when I automatically initiate mubu.process, the segmentation algorithm does not parse the whole audio, and stops somewhere in the middle, and outputs alldone message. If I initiate the mubu.process segmentation twice, then mubu.process parses the whole file.
I think the issue comes from reading a .wav file. I think mubu.process starts processing before mubu container is done with reading. I found a workaround by initiating the mubu.process twice, but I think there is a bug in the reading process. I attached the max patch.
test-segmentation.maxpat (99.9 KB)
I don’t know if this applies to your problem, but in my case the problem was entirely related to redrawing of the views in imubu (perhaps you are using imubu elsewhere in your patch for visualization of the mubu buffer).
My solution was to send the message forceredraw on done (or doneall) from mubu.usual.onseg (in your case mubu.process).
However, I noticed that this will not always redraw the all the views (audio, seg, etc.), but only the foremost one. So to make sure that all the views in imubu redraw correctly everytime I read and process an audio file, I send the message foremost 2, forceredraw, foremost 1, forceredraw.
I hope this is of some help.
Thanks EP for the workaround, I will give it a try. Though, in my case, I am not using any visualization of the buffers, and the patcher doesn’t have any imubu objects.