The opera project that I was working on which included Surtitles , is now up and running successfully. It included 5 surtitle screen areas so surtitles could be placed on a screen above the head of the singing performer, or a singing filmed character on screen..
I started programming this in Glypheo running on the same machine and Syphoned into QLab camera cues, moved to NDI streams running on a second computer, but got totally frustrated whenever a character moved to a different position that I couldn't select a block of surtitles in Glypheo and cut and paste them to a different track/screen position. The new export import features in Glypeo did not help very much as the implementation is very limited..
In the end I went back to doing everything in one QLab workspace and only needed to write half a dozen scripts to be able to import titles from a spreadsheet and convert blocks of cues to Glypheo style transitions (fast fade previous then fastfade up new caption), and a few other utility functions to emulate some of Glypheos other features and to handle batch changes of fonts and sizes etc.
I had about 1000 surtitle cues of which half were automated in sync with video cues and half cued line by line manually to sync with live sections of the performance.
Whenever I edited the video cues, I had a couple of scripts to add or subtract to the prewait times for subsequent cues.
I am now strongly of the opinion that Surtitling in QLab, with a few suitable scripts to hand, is for most uses a much more flexible and more efficient workflow than using a standalone captioning software solution. The sequence recording in QLab is great for syncing surtitles to prerecorded or clicked sections of performances. and the fact that spelling mistakes or small rewrites can instantly be corrected makes it much more flexible than rendering surtitles onto the actual video files.
If you think there is something that standalone Surtitle software can do more efficiently than QLab, I would be interested in hearing about it, to see if I can optimise my scripts to do it more efficiently in QLab!
Mic