I have en entire Bible app in Roman script with the Aeneas timing files created for all the audio. Some of those timing files have been tweaked / corrected. This language is also written in Arabic script, and I would like to re-use those same timing files for the this other Writing System.
I assume I’m going to have to run Aeneas again on all of the audio files in the Arabic script (AS) version of the app as well, so that the Phrases files are created for the Arabic text. Yes, it will also produce timing files, but I assume that as long as the AS SAB project has all the same files, I should pretty much be able to copy and paste the timing files from one project to another, to get all of the corrections.
I think I did something similar before but it was in a Picture Story app. Yes, you should be able to import the timing files. It should work if your punctuation that marks the end of phrases is exactly in the same places. Are you synchronising on the phrase level or verse level? And I don’t think you have to run Aeneas again to create the phrase files. You could test it on some files by exporting it to HTML and watching it in the browser.
I’m pretty sure this is what we did and do with any non-Roman scripts now. IE: Arabic, Hindi. We build the timing files off a Romanized translation, and then use those timing files for the non-roman script and it works - straight out of the box - so to speak. No other processes are needed. If that is not the case, let me know. One of my colleagues has gotten pretty good at working out the kinks for this and may be of some assistance.
Thanks for the input @Friedo and @Dan_Neville. So from what you are telling me, it sounds like the phrases files are only created to produce the timings files? I’ve wondered about why the phrases files are placed (by default) outside the project folder. So I’m wondering how the app actually highlights the right text… Does it keep track of which punctuation is use to split phrases (yes, Friedo, I’m using phrases level breaks), then do the splitting again on the fly, and then just presents the segments one after another? (I.e. It’s not using those phrases at all once the timing file is there?)
But on the other hand, when we modify the timing file manually (see Audio sync with \v 1a, \s \v 1b - #9 by jeff_heath), we’ve been telling people to modify both the phrases file AND the timing file. Is the modification of the phrases file actually not necessary? If it was ONLY used to create the timing file, then ONLY a modification to the timing file would be necessary.
I’m starting to put the pieces together a bit. When the text is processed, it is broken down into the individual segments marked with the segment number, like <div id="T13c" class="txs">. You can see this if you export as HTML then view the source. See this post for more detail:
So the app doesn’t need to split up anything, as the text is already split into its individual segments by SAB when it is written out into the app. I think there is still a question of when and how that splitting happens (as the link above talks about a problem with that splitting), but the phrases files aren’t used, so I believe I can directly copy over the timings files from one project to another that is the same text but just a different writing system.
And in fact, I’ve just done that… and the initial tests seem to be very promising!
One small caveat which I ran into when combining separate apps with timed audio into a single app (two storybook apps that I combined). The two book IDs were the same and the timing files use the book ID in the naming. So my two apps had identically-named timing files. Since they all ended up in the same timings folder in the new app, they overwrote each other. That may possibly apply in your case if you ever use timing files from a different set of books and the IDs conflict.