If you understand Japanese,  do you think it would be possible to create a Japanese translation of the text (in the future)?  Even if the audio is in Japanese,  the menu text, settings,  etc. would still be in English.

Ah, sorry, not nearly well enough to translate something like this! If it were possible to extract the script from the 3DO-J version, I suppose it could be integrated. But it's like the English 3DO version - the spoken text is not subtitled. The only text actually in Japanese are the dialogue selections and the main menu. The rest of the menus and the text are still in English. The rather-thick manual (it's a dual CD case) has Japanese translations (katakana) of the star system names and such.

I had planned to go through the .TS files and retime them by hand. If I get some spare bits this weekend, I'll give it a shot. What I was planning on doing was playing the English and Japanese files side-by-side, and if the Japanese file was shorter, remove the difference, proportionally, from the timing in the file. This wouldn't need to be done for EVERY speech file, but it would still take a bit of time.

Is there any debug function in the game to run through the dialogue scenes so these could be tested?

http://sourceforge.net/p/sc2/uqm/ci/master/tree/tools/synch/ - I can probably cross-build this too, if you want, but it is apparently going to need an extra DLL or two. (It also seemingly doesn't generate subtitles directly and needs a perl script to do that, so you'd need to be able to run perl.)

Haha, as soon as I posted I found the source codes...but yeah, I'd need this compiled for Windows, too. I've no experience with perl though.

I've figured out how the timescript files are structured, at least, I guess I could just do it all roughly by hand, though it could take a bit of time. Of course, that means the subtitles will have to go by quite a bit faster, but it should still be okay.

The speech definitely runs much faster in Japanese. For example, in English, the opening Spathi line is 56 seconds. In Japanese, it's only 25! Part of this is due to how much faster the actor talks, but it looks like the Japanese version plain omits some lines, too.

Thanks!

Linux-to-windows cross-compiled binary of UQM's current abx2wav source - I have no ABX files to test it on, but it seems to run at least:

http://plantmonster.homeip.net/~oldlaptop/abx2wav.exe

This is seemingly a commandline utility which does one file at a time. There's a *nix shellscript included in UQM's source control with it:

Thanks! This did the trick. I've been slowly batch converting these over and successfully tested it within the game.

The problem I've run into now is that, in practically ever case, the Japanese sound files are shorter than their English equivalents. Most of the time it works OK, but there are occasions when the sound file ends before the subtitles do, so it cuts off the rest of them. It looks like I need to edit the timestamp files, and it seems like a program exists that should help me out, but the link is broken. It's mentioned on this page:

http://wiki.uqm.stack.nl/Translations#Tools

Actually, I think I found it here, but again, it's uncompiled:

http://sc2.cvs.sourceforge.net/viewvc/sc2/tools/synch/

If anyone could dig up/compile that tool, it would be a tremendous help. Thanks!

Fun bonus fact: in the English 3DO version, the captain is named Zelnick and the ship is the Vindicator. In comparison, in the Japanese 3DO version, the captain is named Isamu, and the ship is the Kamui.

So I've been toying with an idea to extract all of the voices from the Japanese 3DO version of Star Control II and add them to The Ur-Quan Masters. I bought a copy of the game, made an ISO and extracted all of the voice files, but right now they're all in ABX format. I've seen some references to a program called ABX2WAV but I can only find the source code and not any compiled version. Can anyone supply one for Windows?

After this is done, it should be easy to do a batch convert to change these into OGG files, rename them, zip them up, and hope they match up 1:1 with the voice files in the standard UQM package!