I made a little progress on my Sharkbiter video this week. It has taken longer than I’d hoped. There are far too many things to distract me. I got a new version done, but I’ll need to do at least one revision before I’m done. I’ve done five so far.
I made some substantial changes to the script for the latest version. I also did more with the facial expressions and gestures. I think they make the video more interesting to watch.
The video takes nearly an hour to render, so it is hard to check up on minor changes. I need to create the whole thing before I can see what effect any changes have. The final video comes in at about 9 minutes.
Another problem I’ve run into is that since I need to make work with the whole video at once, I need to keep good notes of what changes I make. The first time I did that, I had to watch the video though a couple of times and modify my script to match. The program doesn’t allow you to output the text as a script.
However, I found out that I could open up the data file the program uses, access the text directly, and transfer it to my script-writing program (Celtx) without much difficulty.
The dialogue is in the .state file, which is actually a compressed .zip file with several files that contain the information for the video. I renamed the .state file as a .zip file and opened it up with Winzip. One of the files inside, document.xml, contains the dialogue.
I opened document.xml in Excel and deleted all the items except the character names and dialogue. Next, I changed the xml code to match the html codes used in the Celtx file. I think I could set up an xml style to do that instead.
The Celtx file, like the .state file, is a compressed .zip file. Two files contain two different versions of the script. script-TF1.html has an older version and script-TF1-1.html contains the most recent version. I just cut and pasted the edited dialogue I got from the document.xml file. Finally I rezipped the file and renamed it.
This approach seemed to work fine. I could open the new file in Celtx and it looked OK. I’m sure that there are some nuances to the files that I haven’t picked up on. I’m sure I’ll find out about them later. I think the conversion could be done the other way too.
If you’ve found an easier way to do this, let me know about it.