So far, transciptions don't exactly line up so well with when people are speaking, and they cut mid sentence. So, I am using the combine/split buttons. However, the split button is really just copy and past, and the combine button combines everything.
Here is what a runon caption looks like it should not be runon like this i wish it were
easier to work with
Here is what a runon caption looks like
It should not be runon like this
I wish it were easier to work with
To fix the above (on two lines)
I would have to split the first line 3 times, edit so there are 3 captions that are different, and then combine the last line with the 3rd split line
It would be much easier to do if the "split" button split at the curson and left anything before it in the current caption, rather than duplicating it.
It would be much easier to do if the "combine" button combined anything before the cursor with the previous caption, and left the rest in the current caption.
Maybe "add text before the cursor to last caption line", "add text after the cursor to next caption line" and "split current caption at cursor" would be more intuitive, faster, and less error prone?
There has been a lot of discussion about how to structure this type of operation, where the cursor (CTI/current time indicator) is used as the point to divide text. Your description presents good options. Numbering them: 1) "add text before the cursor to last caption line", 2) "add text after the cursor to next caption line" and 3) "split current caption at cursor"
There are keyboard shortcuts for the split and merge operations. They do not have assigned keystrokes by default. For now, they make these operations much easier than clicking in the Text Panel. Their descriptions are:
"Split caption segment under CTI" which duplicates the text in both segments, and
"Merge selected caption segments" which adds all text from both segments.
The timecode part of CTI placement is already respected: when a caption segment is split, the CTI position is used for the new end/start timecodes.
I think the feature request above is your #3, and that alone would be a major step forward. Change Split to have an option for the a new behavior, or just change it. Does anyone ever want to duplicate the entire caption text to both segments? The new behavior might be:
"Split caption TEXT under CTI" so text before the cursor goes to the new, earlier caption segment and text on or after the cursor goes to the new later caption segment. And, as currently, the CTI position would determine the new end/start timecode.
Your # 1 could then be accomplished by a second operation where you merge the new prior segment with the one before that, and #2, the new later segment with the one after that.