C&C TD - Italian translation

[Nyerguds]

About "Music Factory", I admit my ignorance about it. I thought that label was a building name, i.e. a factory producing musical instruments. What exactly is "Music Factory" in the game?

These are the names of the civilian buildings. The game normally doesn't show them (instead it shows "Civilian Building"), and because of that, the Westwood people gave them funny names. They can be enabled with the option TrueNames in rules.ini

 

As for the music tracks, no, that's no problem at all. It's just, most translations don't touch them.

 

If all text translation is now OK, I have a new question: how can I add subtitles to movies seen in the game? I see a "subtitles=0" option in conquer.ini... so, can a subtitles text file be used? Or that option has another purpose, or it was never used? Or I must physically edit frames, and add subtitles manually to them? If so, how?

The subtitles option in the game has no effect. It is possible to edit the videos and add them in, but I'm afraid there's an extra factor involved, since the videos are 256 colour.

 

Last question about this... where can I find a transcript for every movie (including the long introductory one)?

There is a thread on CnCNZ where they did the same thing for spanish. I think the thread has the transcripts  of all C&C videos.

http://forums.cncnz.com/topic/17046-translating-td-videos-to-spanish/

 

After this, the last challenge will be in-game speech... it must be re-dubbed, of course. Options are two: take Italian audio from other games, or using an Italian speech synthesis engine. But... only after having subtitled movies.

If there are italian versions of TS or C&C3 and you can get me all EVA sound files of these, I could probably make an EVA voice set if you give me the italian translations for all the voices in speech.mix. I've done that for the Spanish one as well.

[maancheno]

Ok, the speech.mix for Italian is complete. Obviously, for lack of material, some translations are "almost exact"... but not perfectly exact. But I've managed to translate ALL audio samples, except Kane's speech (sigh!). Next step will be talkeng.mix, and all the "audio" part will be finished.

speecita.mix

[FunkyFr3sh]

wow nice, from where you have these sounds ?

[maancheno]

The last audio part: talkita.mix (the translated version of talkeng.mix), with the Italian unit voices.

 

I've taken all audio material from Tiberium Wars.

 

Now... time for video subtitles!

talkita.mix

[maancheno]

I'm starting video subtitling task. It's heavier than it seems.

 

As a starting point, I'm looking for English transcripts of videos. I've found a document from another forum about development of 1.06c patch, http://forums.cncnz.com/topic/13910-106c-development-topic/page-36#entry211071, which contains a document named "Spanish subtitles and frame numbers.docx", written by Nyerguds, which contains most of transcripts, but in Spanish. Does a version of this document exist with original English transcripts? Because translating by hearing, or from another translation, would be error prone, and I know English but not Spanish...

[Nyerguds]

That's pretty old, I think. The other thread I linked to contains all English transcripts, you just have to browse through a few pages to get them all.

 

Do you want me to compile it all into one document?

[maancheno]

Yes, please... I need ALL transcripts, not only those mentioned in that thread; for example, I can't find retro.vqa and intro2.vqa in it...

[Nyerguds]

Hmm. It seems the uppercase "à" character can be used if you switch to the Portuguese text encoding (DOS-860). I made fonts for Portuguese, so all that would need to be done is including them and converting all text.

[Nyerguds]

The speech mix file you attached doesn't work; all the audio files in it are .wav instead of .aud format.

[maancheno]

The speech mix file you attached doesn't work; all the audio files in it are .wav instead of .aud format.

 

Whoops 

You're right, I forgot the last step.

 

 

 

speecita.mix

[maancheno]

I'm trying to translate RETRO.VQA. I've found a Youtube video, C&C "Rescue Mission" cutscene in which a transcript exists. But I don't understand some parts of it.

The transcript is this:

 

I don't think you understand what's happening here.

If you don't spill it, I'm gonna (=going to) spill it!

You! All over the damned floor, now!

Give me everything I want, and give it to me now, or it's over for you! it's...

That's really funny, tough guy, really funny! Here you go!

 

Ok, the literal translation is not difficult, but:

What does "If you don't spill it, I'm gonna spill it" mean? Or it actually is "If you don't spit it, I'm gonna make you spit it", or something similar?

Is "You! All over the damned floor, now!" exact? What it's referred to? Not to the victim, because he is still sitting after this sentence...

[Nyerguds]

Not sure if that's worth translating... it's just a rough interrogation and execution of a prisoner, is all.

 

As for the "if you don't spill it", the English expression "spilling the beans" means revealing a secret. The "I'm gonna spill it", of course, refers to spilling the prisoner's blood (or brains, here ), AKA, killing him, and to stress it even further he even explains that what he's going to spill is "You! All over the damned floor".

 

Which he does, too. And on the wall. and the window

[maancheno]

Mmmmh... so hard to express it using the same spill=spill wording. And idioms...

 

I had to translate it like this: "Se non parli, qui ti ci spalmo!" ("If you don't speak, I'll splatter you here!")

 

But... I also made an improved version of text_ita.ini, with Portuguese encoding, and the lang_ita.mix containing it. With this files, Italian translation of texts and audio is complete... except bugfixes, if needed.

 

I'm still working on video subtitling, using this algorithm:

- XCC Mixer: convert VQA to list of PNGs and WAV

- VirtualDub: convert PNGs + WAV into Old Format AVI (with 15 frames per second)

- Subtitle Workshop: create .SRT subtitles for video

- VirtualDub - TextSub plugin: add subtitles to video

- RADVideoTools: convert subtitled video to PCXs

- VQA Encoder - old-enc.exe, ugm2vqa.exe: convert PCXs + WAV to VQA

 

Keeping in mind original resolution, results are not so bad. For example, I've attached resulting GENERIC.VQA to this post (it's the smallest one I've made till now )

GENERIC.vqa..rar

text_ita.ini.rar

lang_ita.mix

[Nyerguds]

Heh, wow.

 

Not sure if you saw the discussions with Gordan in the Spanish videos thread, but one thing to note is that when converting to PCX, you can technically split the video into scenes which each have their own palettes, to prevent getting too much quality loss in videos that have really different colours in different scenes. The catch is that in the existing VQA encoder, this HAS to happen on a multiple of the configured key interval setting. So if it's set on 8, as it normally is, these palette changes can only happen at the end of every batch of 8 frames.

 

You can do this by converting with RAD from video to high colour PNG frames, then separating those into different folders, and then using RAD tools again on each folder of frames to convert them to paletted PCX. The only disadvantage is that all new frames will be numbered from 000 again. This can be fixed with this simple Copy/Rename tool,

[Nyerguds]

Well, I finally got testing on this again! There were still a few errors... the Desert icons weren't in the Desert palette, and the speecita.mix crashed the game because it was created in TS mix format instead of TD format.

 

All of that is fixed now, and I generated a new conquer.ita from the ini version of the strings file, and included the Portuguese fonts. The final result is attached.

 

I could make an installer out of this, if you translate the installer messages:

 

WelcomeLabel1=Welcome to the C&C95 Italian Language Pack Setup Wizard!

WelcomeLabel2=This program will install the Italian language pack on your C&C.

AppName=Command & Conquer Gold 1.06c Italian language addon

nogame=ERROR: The game was not found in the selected directory.%nPlease select the directory where Command & Conquer is installed.

EnableLanguage=Enable Italian language

 

[edit]

 

I found a missing string in the strings file:

733=Command & Conquer is unable to detect your mouse driver

 

You seem to have translated the old version of the string... which said "unable to set video mode". This error is no longer in the strings file, because it is always shown before the language files are loaded anyway (which gave an empty error message box).

addon_italian.rar

[maancheno]

Tests... they're never enough!

 

I've corrected the INI file with missing string, I've attached the result. I've also replaced the abbreviation of billion,"BLN", with its Italian translation, "MLD" (miliardo).

 

For installer strings, their translation is this:

 

WelcomeLabel1=Benvenuti nella procedura di installazione della traduzione in italiano per C&C95!

WelcomeLabel2=Questo programma installerà la traduzione in italiano dell'applicazione C&C.

AppName=Componente aggiuntivo italiano per Command & Conquer Gold 1.06c

nogame=ERRORE: Il gioco non è stato trovato nella cartella selezionata.%nSelezionare la cartella in cui è installato Command & Conquer.

EnableLanguage=Abilita lingua italiana

 

I'm still subtitling videos... it requires a lot of time... and experimentation. A solution could be modify the application, by overlaying text on original video screen, and introducing a INI-like file format for subtitles, but I suppose it's not so easy as it seems. But also I'm a programmer, and I could give it a try. It would be a great challenge. Or perhaps, a Mission: Impossible.

text_ita.ini.rar

[Nyerguds]

I've looked into that, but writing subtitles over the video play seems horribly complicated... especially with each scene having its own colour palette. I actually designed an ini format for this once in case I could implement it, and it also required a palette index to use for the colour.

 

I should ask iran how fonts actually use colours, though...

[maancheno]

I thought it was simpler... like adding a new transparent "label" over the screen without altering video data...

 

Another solution could be creating an external video subtitler, capable of editing video data (so adding white pixels for text) without decoding and re-encoding video. The XCC Mixer tool is already capable of extracting frames from videos, so it's not impossible... and perhaps I could try. It could become a translation-specific tool, for example.

 

But... perhaps this problem is simpler: where can I find ALL the English transcripts of videos? I've already downloaded TDTranscribed.txt and TDTranscribedII.txt from http://forums.cncnz.com/topic/17046-translating-td-videos-to-spanish/ about Spanish translation, but they contain only some of them... for example, other videos like nod2.vqa are missing.

[Nyerguds]

I thought it was simpler... like adding a new transparent "label" over the screen without altering video data...

That would indeed be pretty much how it'd need to be done, though, I think C&C doesn't need any 'label' objects to write text on screen. Problem is, the game is still 256 colour, and the videos use that full scope of 256 colours to get good video quality. Where the normal game makes sure certain colours are always the same on all palettes the game uses, like white, black, certain greens etc, the videos don't do this at all. This means you'd have to figure out, per video, and even per scene with a different palette per video, which colour index to use. I'd also probably have to rewrite the text to the screen buffer after every single single frame load.

 

At the moment, I haven't really experimented with this since I'm not too sure how the loading, using and colouring of fonts works.

[maancheno]

Ok, so in-game subtitling seems to be hard.

 

Instead, I'm trying the other way... writing an external subtitler. My ideas are essentially two:

• Writing a "direct" subtitler, capable of writing white pixels (shaped as text) into frame data. It seems to be VERY difficult... I'm still stuck at frame data decoding, and palette problem will still be a problem. It will be a VERY long work. But quality loss of original video will be ZERO.
  
• Writing an "indirect" subtitler, which (using existing command-line tools) converts VQA into PCX+WAV data, writes text on PCX frames (e.g. taking it from an .SRT file and converting timing to 1/15 second intervals), and rebuilds the VQA from modified PCX+WAV. It seems much simpler, but a bit quality loss will occur. But not so much as converting VQA -> PNG -> AVI -> subtitled AVI -> PCX -> VQA like I'm doing now. And there's a problem: I've already found a command-line VQA encoder, but not a decoder. I can perhaps extract it from XCC Mixer...
  

 

Both approaches could have a great advantage: they require a .SRT file, which is a text file. If a list of English .SRT files could exist, a translator could take them, translate their content, and apply them to videos, like .INI files now.

 

P.S.: I'm still writing Italian .SRT files, but there's always my problem with spoken English... for example, what's saying Kane in this video (Nod mission 12, Capture the Communications Center) C&C Tiberian Dawn - Nod Mission 12 - Capture Communications Center when he says "As you ???????????, GDI has begun field testing his satellite based ion cannon despite your previous success" ?

[Nyerguds]

You can't write it into the frame data, since the format is encoded in blocks of frames... it'll need to be re-encoded with Gordan's converter anyway.

 

As for the video, what he says is

"As you are no doubt aware, GDI has begun field testing its satellite-based ion cannon despite your previous success."

 

[maancheno]

Then, indirect way will be the only one. Now, first and last steps are already OK:

• decoding, using a (working) command-line tool extracted from XCC Mixer's C++ source code, which extracts .PCX frames and .WAV audio;
  
• encoding, using command-line version of Gordan's VQA Encoder, which takes .PCX and .WAV input and builds final .VQA file.
  

The rest to be written is... the actual subtitler, which will take a .SRT file and a sequence of .PCX files as input, to produce a new list of subtitled .PCX frames. I'm choosing C# as programming language, because writing GUi will be easier, as image manipulation: I've already found a way to write text on images, so the hard part will be only converting subtitle time intervals to 1/15 second rate, and computing exact position of text on frames. First goal is "batch" subtitling, then an "interative" mode could be added, like Subtitle Workshop does... but this will be hard.

 

Also data input for subtitler is almost complete: I've written all Italian .SRT files for videos... at last. And, before it, a text file with all original English voice transcripts, based on data from

• http://forums.cncnz.com/topic/17046-translating-td-videos-to-spanish/, the main source
  
• the old document with Spanish transiations (re-translated to English and refined by hearing)
  
• some help from Youtube automatic subtitler
  
• last but not least... this forum!
  

 

The full English transctipt is in the file attached. Some videos are missing, but only because the part to be subtitled is not in speech, but written as text on frames, like in VISOR.VQA, OBEL.VQA and so on.

There are only two "holes" in this file, marked by ???, which I could not transcribe... can you help me?

MOVIES_SPEECH_TRANSCRIPT.rar

[Pigeon]

From my personal experience (polish translation) i may tell you, that far way easier it would be to add voice actors, because all You would have to do, is to mute vocals and add new ones. Problem is to find any volunteers for this

[Nyerguds]

Uhh... you got Polish dubs of the C&C1 videos? (if so, gimme!)

 

By the way, you're wrong there. Dubbing isn't that easy; it's really hard to take away the voices while retaining the background ambient sounds. Without those, the dub will sound terribly cold and blank.

[Pigeon]

Uhh... you got Polish dubs of the C&C1 videos? (if so, gimme!)

 

By the way, you're wrong there. Dubbing isn't that easy; it's really hard to take away the voices while retaining the background ambient sounds. Without those, the dub will sound terribly cold and blank.

 

I know some "magicians" (anime fans), who dubs japs animations quite good. Obviously, low-cost mics and missing talent can't be compared to originals, but it's enough for fan dub.

http://nanokarrin.pl/portal-2-end-credits-song/

[iNSOMNIA] "Lamy w czapkach (Odc. 3)" DUB PL

[iNSOMNIA] "Na g?upi? ?mier?" DUB PL

Fullmetal Alchemist: Brotherhood - Opening 1 [polish fandub]

 

Maybe something more comperable:

[iNSOMNIA] "Gor?ce Lato" DUB PL vs ????_HOT SUMMER_MUSIC VIDEO

 

From what i know, erasing vocal is not possible, but it is possible to "hide" it. Results are quite surprising... despite that i am more then sure they worked on lossy mp3 format.

The main issue in C&C dubbing is two-way lossy convertion VQA-AVI and back... and enough volunteers for dubbing. I was even considering back in 2002/2003 to add a reader to C&C cut-scenes - what is popular in postsocialist countries (and even now) - but i didn't had enough knowledge for this.

Probably i would be able to try dub a one random cut-scene from C&C (for test), because i have good contact with both polish fun-dub studios, but right know i don't have enough time to bother with this (exams are coming). Moreover, i don't want to make any kind of promises, because as Uncle Ben said "with great power comes great responsibility

 

What concerns polish transcript - i don't have it anymore. I've erased everything right after i've abandoned this project. After all i didn't know, that i would work on it ever after... and i didn't Translation of C&C95 doesn't count!