Qhimm.com Forums
Project forums => FF7Voice => Topic started by: ff7heart on 2008-05-24 09:00:00
-
I can guess that it would be a mammoth project if anyone ever thought about doing it! But yeah, just wondering if you've ever heard of anyone even mention such a project before? I know we all have the possibility of a remake to look forward to, but in any case, that seems a long way off!
Would such a project be possible? I mean, technically? I'm new to the entire mod/patch etc scene, but would it be possible to include voice overs? Hmmm... I wonder!
If I wasn't a 16 year old kid with exam-overdrive to contend with, I might have actually thought about starting this up.
It would have collapsed within hours, of course, but hey, I'm the eternal optimist :]
-
Wouldn't that be imposiblle?
What with FFVII havign all it's files already, wouldn't it require creating new files/archives, and packing them into the games directory?
I just ask because I would think that would make the game incapable of running.
Something no one would surely EVER want. :-)
-
If you could could make it run like ff7 music, outside the actual game, where it just intercepts, say, the level data and uses that to select what file to use, it should be semi-possible. Not realistic, but possible.
-
It theoretically could work I'm sure. But there would most likely be some terrible loading times.
-
well its quite possible all you really have to do is get the map files and get them to trigger a noise and you make that noise the speech.
i think meteor the program that is would be capable of adding the code to every scene, can be done but you would need lots of time some good voice actors and it would more then double the size of FF7 on your hard drive, its a case of too much effort for not enough gain, don't get me wrong if someone did it i would probably download it but i wouldn't be the one to make it.
-
I'm a Voice Actor and will voice some of the Characters for free if anybody wishes to attempt this.
I can be contacted either by E-mail (Undernet01(at)Shaw(dot)ca) or by MSN (Mesden84(at)Hotmail(dot)com).
My voice wouldn't be a fit for Sephiroth, Barret, Cid, Vincent, Cait Sith, Tseng, Etc... Their voices are either too deep or too aged... But I'd be able to ace Cloud, Zack, Reno, and Rufus. Send me a shout if anyone wishes to get this started.
-
My voice is rather deep, I reckon I could pull of Sephy. ;-)
-
Hahaha.
Nice idea.
I'd also be able to volunteer to voice given a timetable to work with. My voice is not a deep one either and I'm by no means professional but I could gladly do NPC's and such or whatever you guys would see fit.
Warning though, I speak English perfectly but I do have an accent as I'm not a native English speaker.
But before getting into all of this, I think we should ask ourselves this question : would an amateur speech patch (even if some professionals are on board, not everyone would be) actually improve the game or make it worse ? As for myself, wrong deliveries, wrong emotions in a given scene and wrong casting totally takes me out of a game. The first Grandia and Tales of Eternia are good examples of this. I'm asking myself if these two games would have been better off without voice overs as an example.
As for the technical side of it, in order to reduce the size, MP3's would be a logical choice and ff7music could most likely be used to handle the playback if Ficedula agrees with it as, from what I understand, it already can intercept the calls the game makes and start its own sound engine on cue so I think it would be the easiest way to make this work. It would need some modifying obviously though.
Going on a limb here as I'm no programmer but I would think that the game engine should make a call referencing to a certain part of the game script in order to pull the right text (in scene.bin if I'm not mistaken) so it would be likely that these calls could be intercepted to start the right sound bite through whatever program would be used for playback without modifying ff7.exe at all (Ã la ff7music). The trick is going to be to find out what are the actual calls the program makes for text and interpret them to actual events (so in other words, find out how the game ID's event to show the right text so the speech player could intercept those and start the proper MP3). A mapping of the text might already exist though since I know people have been playing with the dialog before for various patches. There might also be info somewhere as to how the game fetches text.
As such, using ff7music could be beneficial to use as it already starts ff7 in a way to intercept its calls, already has a list system to associate sound files to events in the game and already has a sound output system. Obviously, this is Ficedula's program so he would have to agree (and most likely lend a hand to the programmers understanding his code) so it might not be possible to use his program but that would be the #1 solution in my book (or a similar program if not possible ff7music).
Another issue would be that the text is hardcoded into the FMV's so for these it would actually be necessary to code the calls for the speech into the game. That's the only part I see that actually requires modifying of the game. Unless, again, the calls made by the game to start a fmv can be intercepted to determine what fmv it is starting and put in a timer to play the required speech file. That implementation could easily be off-cue though so that might not be the best way to go at it.
Overall, I'm thinking the big parts could just be to add a layer on top so it should not be impossible. Obviously, I'm no programmer and I could be totally in left field on this as this is just educated guesses from what I've read over the years, so if someone more informed reads this, feel free to correct me.
-
Hahaha.
Nice idea.
I'd also be able to volunteer to voice given a timetable to work with. My voice is not a deep one either and I'm by no means professional but I could gladly do NPC's and such or whatever you guys would see fit.
Warning though, I speak English perfectly but I do have an accent as I'm not a native English speaker.
But before getting into all of this, I think we should ask ourselves this question : would an amateur speech patch (even if some professionals are on board, not everyone would be) actually improve the game or make it worse ? As for myself, wrong deliveries, wrong emotions in a given scene and wrong casting totally takes me out of a game. The first Grandia and Tales of Eternia are good examples of this. I'm asking myself if these two games would have been better off without voice overs as an example.
As for the technical side of it, in order to reduce the size, MP3's would be a logical choice and ff7music could most likely be used to handle the playback if Ficedula agrees with it as, from what I understand, it already can intercept the calls the game makes and start its own sound engine on cue so I think it would be the easiest way to make this work. It would need some modifying obviously though.
Going on a limb here as I'm no programmer but I would think that the game engine should make a call referencing to a certain part of the game script in order to pull the right text (in scene.bin if I'm not mistaken) so it would be likely that these calls could be intercepted to start the right sound bite through whatever program would be used for playback without modifying ff7.exe at all (Ã la ff7music). The trick is going to be to find out what are the actual calls the program makes for text and interpret them to actual events (so in other words, find out how the game ID's event to show the right text so the speech player could intercept those and start the proper MP3). A mapping of the text might already exist though since I know people have been playing with the dialog before for various patches. There might also be info somewhere as to how the game fetches text.
As such, using ff7music could be beneficial to use as it already starts ff7 in a way to intercept its calls, already has a list system to associate sound files to events in the game and already has a sound output system. Obviously, this is Ficedula's program so he would have to agree (and most likely lend a hand to the programmers understanding his code) so it might not be possible to use his program but that would be the #1 solution in my book (or a similar program if not possible ff7music).
Another issue would be that the text is hardcoded into the FMV's so for these it would actually be necessary to code the calls for the speech into the game. That's the only part I see that actually requires modifying of the game. Unless, again, the calls made by the game to start a fmv can be intercepted to determine what fmv it is starting and put in a timer to play the required speech file. That implementation could easily be off-cue though so that might not be the best way to go at it.
Overall, I'm thinking the big parts could just be to add a layer on top so it should not be impossible. Obviously, I'm no programmer and I could be totally in left field on this as this is just educated guesses from what I've read over the years, so if someone more informed reads this, feel free to correct me.
It would be up to a Director to get it correct.
-
Well, I definitely support this idea and would love to be involved as a Voice Actor.
-
There's plenty of good voice talent who'd probably be willing to work on a project like this. The problem is actually inserting them into the game. You could probably look to other communities for help on voice acting, so you'd get the best results. pixietricks from OCRemix came to mind immediately as someone who could probably voice a female character well, and be willing to do such a project. But like all projects, you need to show you can pull it off and have some charisma to bring in the talent.
-
There's plenty of good voice talent who'd probably be willing to work on a project like this. The problem is actually inserting them into the game. You could probably look to other communities for help on voice acting, so you'd get the best results. pixietricks from OCRemix came to mind immediately as someone who could probably voice a female character well, and be willing to do such a project. But like all projects, you need to show you can pull it off and have some charisma to bring in the talent.
There's plenty of good Voice Over talent, Yes. The problem is you can't get a Voice-Over Actor who is Union-based, and most talents would want money for a project this big. You'd need to find people who'd be willing to do it independently. If someone would like to program this, I have sources for individuals who would love to Voice-Over for a project like this, and I definitely have experience as a Voice-Over to be a Voice Over Director to make this a success. (Just throwing ideas out there.)
-
I'd do it, hell I've had practice at it. When I played the game with my friend watching we did a 2 actor voice over of the whole game. I don't think my voice would be right for the deeper voiced characters, and isn't good enough for cid, but like Mesden I could do like Zack, Reno, maybe Tseng and Rude, and rufus as well.
-
I still stand by I could do Seph, as my voice is deep, I talk in a good enough pitch, hey presto, Sephy.
And, I'd do it for free on weekends. :-)
-
Hi all
I remember seeing a snes game, that got Voice-Over mod that was a RPG
I would be willing to do some Voice-Over's
the in-game movies can be done right now
But before we all start saying, i want this part or that part.
That we try and get a Voice in-game
If it truly can be done
You would need
Some people to get the game to take the Voice's
Some girls to Voice-Over
Some deep Voice people
I think that deep Voice's are good for Cid and Barret
I feel that Sephiroth would sound mad when speaking normally then sound crazy later on
some-thing like Jack Nicholson in the The Shining or batman as the joker
-
I don't think Sephiroth should sound anything like Jack Nicholson.
-
LOL same here
I was talking about the way he start off normal, in the shining then by the end. He giving off that crazy feeling
tho that might take some doing for some-one who never did Voice-over's before
What i think im getting at is we need some people. Who can make you feel that they are there in-game counter parts
-
Aside from the issue of how good such a mod would be, I'm afraid to say that FF7Music really wouldn't be any use for this. It doesn't monitor what the game is doing very closely at all - all it really knows is when the music track changes. (The same technique can be used to find out when the location changes - but not anything more detailed, like which line of dialogue is showing on the screen...)
This isn't to say that couldn't be done; it could, but it would require a completely different method of hooking into the game.
-
The Meteor topic (http://forums.qhimm.com/index.php?topic=6020.25) has a few posts on this.
-
Thanks for the feedback Ficedula.
I've been trying to understand how the game dialog works to see if my theory was correct on how to go at it (which is proving kind of hard since I have no programming skills ... at all) and reading Gears, I see you contributed some info to the Dialog and Event Field Format. Would you have any suggestion as to how this could be handled ?
From what I could gather so far, the Flevel.lgp file will contain the info for each screen and will command the engine to go look into the dialog table to fetch the right text.
Also, the two following opcodes are the one calling up the correct dialog according to the wiki :
0x41 MPARA
[0x41][bank_for_variable_id/bank_for_value][window_id][variable_id][value]
Writes bank_for_value to 800E4214 memory bank variable array (size 0x08 0x01). This value will be used when we will copy field dialog to window string.
Writes value to 800E4D48 offsets for variable array (size 0x10 0x02). This value will be used when we will copy field dialog to window string.
0x42 MPRA2
[0x42][bank_for_variable_id/bank_for_value][window_id][variable_id][ [value] ]
Writes bank_for_value to 800E4214 memory bank variable array (size 0x08 0x01). This value will be used when we will copy field dialog to window string.
Writes value to 800E4D48 offsets for variable array (size 0x10 0x02). This value will be used when we will copy field dialog to window string.
So basically, we would need to watch these specific memory locations and detect the values, reference them to the dialog table which we had mapped previously and send the correct ID to a playback software which would launch the associated sound bite ?
After that, we would need to monitor the following opcode as well in case the user closes the text window before the end of the dialog to stop the sound bite and go on to the next one :
0x2E WCLS
[0x2E][window_id]
If window opened call "set_state_to_close" and then "manage_window_states" to close window.
This opcode called until window closed (we can call this opcode from any entity).
Again, I'm no programmer. I'm just trying to understand a way to do this from a technical standpoint with basically no previous point of reference. Would this make sense or am I totally wrong about this ?
-
This is a fun idea, but I fear the chances of finding voice talent that will improve the FF7 experience is going to be very difficult.
My suggestion is to recruit the voice over actors but only add voices to the more critical/important scenes.
There is a ton of dialog, which means a lot of people's free time and makes it more difficult to focus on quality and consistency with each scene. Either way good luck to everyone who gets involved.
-
Hi all
I remember seeing a snes game, that got Voice-Over mod that was a RPG
I would be willing to do some Voice-Over's
the in-game movies can be done right now
But before we all start saying, i want this part or that part.
That we try and get a Voice in-game
If it truly can be done
You would need
Some people to get the game to take the Voice's
Some girls to Voice-Over
Some deep Voice people
I think that deep Voice's are good for Cid and Barret
I feel that Sephiroth would sound mad when speaking normally then sound crazy later on
some-thing like Jack Nicholson in the The Shining or batman as the joker
Don't forget Cait Sith's Accent... I still think that was the wrong choice for Square to make...
Guys, the people in the Programming Forums brought up a good point. We would need to decide on the Cast BEFORE the Programming were to take place. To do that, we would need to Audition, choose our Voice-Over Actors, and present them to the Forums. Once we have a solid idea of who our Voice-Over Actors would be, it would be a LOT easier for a Programmer to say, "Alright, I'll be a part of this." With everything up in the air like this, it would be next to impossible for a Programmer to commit himself/herself to a huge project.
There are a lot of Programmers on these Forums who are just waiting to become involved with something like this, so it can happen if the willingness to proceed is strong enough from both sides.
Therefore, I'm throwing out a Casting Call to the Qhimm Community. This is to find Talents who are able to *voluntarily* commit themselves to voice-over a character role. Once we KNOW we have the Talents, I'm nearly certain a Programmer would step forward to implement this idea into the Game. I will make a new thread in regards to this, and post the instructions in there.
-
Maybe contact this guy Crisis Core Fandub. (http://youtube.com/watch?v=tNpEfHFmkMQ)
-
I hope Ficedula gets involved as a Voice Over :)
-
So you guys are SURE you can make this work?
-
So you guys are SURE you can make this work?
I have a lot of knowledge in the area of Voice-Over work. Enough of it that I have no problems with being the Voice-Over Director for this project, presuming nobody else objects. Programmers on the Qhimm Forums need to see commitment, and if we can find a talented bunch of Voice-Overs, that will be a good start. People are welcome to submit their Audition via E-mail between 11:59 tonight, up until mid-June -- Though I URGE anybody Auditioning to connect to the Ventrilo Server on June 7th / June 8th. This will play a large, deciding factor on whether or not we have the commitment to proceed. So refer your Facebook Friends, MySpace Friends, MSN Friends, Real Life Friends, Family, Relatives, Etc. Anyone you feel would have the voice that actually fits a Character in FF7.
Information in regards to this may be found in the Casting Call thread.
-
Thanks for the feedback Ficedula.
I've been trying to understand how the game dialog works to see if my theory was correct on how to go at it (which is proving kind of hard since I have no programming skills ... at all) and reading Gears, I see you contributed some info to the Dialog and Event Field Format. Would you have any suggestion as to how this could be handled ?
I have one or two ideas about how it could work. I expect there's probably a number of approaches you could take to hook into the game and intercept the text display, but really technical investigation would be required to see how plausible any possible approach would be. The two that come to mind immediately are modifying the field files to try and play the sounds in-game, or writing an external utility (like FF7Music) that hooked events but did all the sound processing in a separate application. My personal feeling is that the second would be easier.
I'm willing to take a look at how possible something like this would be, but first I'll have to get FF7 working on my current computer!
I hope Ficedula gets involved as a Voice Over :)
I'd be happy to contribute my smoooooth British accent if the project got off the ground, but I don't know I'd be suitable for a main character. If nothing else, the time involved in recording a large amount of dialogue is probably more than I could commit to at the moment.
-
The first option would work better if implemented :P MP3 probably won't get a good compression ratio when put into LGP files, so it might as well be put into a custom container. New modified version of window opcode could be implemented, that would take an extra argument of the MP3 to be played - that would require modifying the flevel files, but once it's done you don't have to peek any process for changes + you'd get a better responsivness.
-
First off, I'd like to say thanks to everybody who's taken the time to consider this so far. It's really great to see people showing interest in this just like I am , especially since two people I would describe as a master of the add-on while the other as the master of fixed/adding to the existing engine from within are among the group.
Also, I've been a member of this board for a few years now and I've mostly taken from this community. I'd like to give back a little so I'd like to offer you guys my help on the technical side but as I've stated before I'm starting from scratch so I don't know how much help I'm going to be.
As Ficedula mentioned, I do believe the first step would be to probe on how the game truly handles dialog events. Something in my grasp right now is to do research so I'll do a research tonight and post the results of my findings on what is currently known on dialog events for quick reference and to further my understanding.
edit : summarized what I was saying.
-
Thanks for the feedback Ficedula.
I've been trying to understand how the game dialog works to see if my theory was correct on how to go at it (which is proving kind of hard since I have no programming skills ... at all) and reading Gears, I see you contributed some info to the Dialog and Event Field Format. Would you have any suggestion as to how this could be handled ?
I have one or two ideas about how it could work. I expect there's probably a number of approaches you could take to hook into the game and intercept the text display, but really technical investigation would be required to see how plausible any possible approach would be. The two that come to mind immediately are modifying the field files to try and play the sounds in-game, or writing an external utility (like FF7Music) that hooked events but did all the sound processing in a separate application. My personal feeling is that the second would be easier.
I'm willing to take a look at how possible something like this would be, but first I'll have to get FF7 working on my current computer!
I hope Ficedula gets involved as a Voice Over :)
I'd be happy to contribute my smoooooth British accent if the project got off the ground, but I don't know I'd be suitable for a main character. If nothing else, the time involved in recording a large amount of dialogue is probably more than I could commit to at the moment.
Just to clear this up, I realize it's hard to fit a british accent into the Game, but Reeve would be a good choice for a british accent. If you want to voice one of the main characters, you would have to learn more of a north american dialect :P
-
Quick progress report on my end for people interested.
I've done a search in technical and in q-gears on every post containing "dialog". I've also read the relevant parts of the wiki. This obviously was massive amounts of info (not all of it relevant for this project) and I need time to let all this set in and I also would like to look at a field file inside flevel.lgp
I can however say right now that I was wrong in that the 0x41 MPARA and 0x42 MPAR2 opcodes are for showing dialog. The correct one seems to be 0x40 MESSAGE.
-
I'll wrap up a quick example of how it could be done by modifying ff7.exe (placing hooks actully). Now I see that there is no need to invent a new opcode or to modify flevel.lgp.
-
Yep, a hook is the way to go; not too difficult to implement. Proof of concept;
http://www.sylphds.net/ev2/img/dialogue1.png
Would need more work to implement sound playback on demand, etc, but at least we know it's fairly trivial to intercept the text as it gets displayed on the screen.
(Ignore the corrupted graphics, getting it working on my Mac is still a bit of a work in progress!)
-
And that quick example I was talking about:
Link (http://dziugo.republika.pl/misc/VOP.rar)
Instructions:
- Place a "sample.wav" file into FF7 folder
- Run FF7 (1.02 patch, English version)
- Run "DLLLoad.exe" from the Bin folder
From this point, whenever you talk to someone, a sample file will be played, and the info about the dialog will be put into debug.txt file in FF7 directory. If there is no sample file, the default Windows sound will be used. If you can't hear anything, make sure that the sound is not muted in FF7 (FF7Music users).
The DLL creates the hooks, and has 2 functions called when the dialog is starting to get displayed, and when it disappears from the screen. Functions are feeded with the ID of current location, and the Dialog ID (in the flevel.lgp files). If the dialog contains multiple screens, it's treated as one, but that could be fixed by (for example) counting the "New screen" tags (then the function would get the third parameter, no# of the screen in the dialog).
Oh, and the source is included only for the DLL, as the Loader isn't actually needed (you can use any dll loader).
-
Great job guys!
Didn't think we'd see results so quickly.
I however had a few questions.
First one, Dziugo, I tried to run your tool to see if I could gather more info about how the game handles dialog but when I start dllloader, i get a side-by-side error. Obviously I know this isn't even alpha yet and this is to be expected but I figured you'd like to know. I use Vista and it was in the right directories.
Second one, Ficedula, I tried to take a look in Cosmo to some of the flevel.lgp files but Cosmo oddly opens outside my screen ? I can bring up the options with alt but can't see the actually main folder.
Third one, I was hoping to have a theory to suggest here but since I couldn't take a look at the files as I would have liked I don't really have much to say. How would you guys think we could identify which dialog line is being called (basically ID it). I mean, we can show the text right now, but can we uniquely identify which text is shown so we can associate that identifier with a file to play it ? My guess from what I know would be that we'd need to identify the right field file in use but apart from that how could we tell which dialog is being called up by the game ?
-
Bah, the old Cosmo off screen bug...easiest solution is to run Cosmo then hit Alt-Space to bring up the system menu (don't worry if you can't see it), and hit 'M' for 'move', then use arrow keys & mouse to bring it back on screen.
Third one, I was hoping to have a theory to suggest here but since I couldn't take a look at the files as I would have liked I don't really have much to say. How would you guys think we could identify which dialog line is being called (basically ID it). I mean, we can show the text right now, but can we uniquely identify which text is shown so we can associate that identifier with a file to play it ? My guess from what I know would be that we'd need to identify the right field file in use but apart from that how could we tell which dialog is being called up by the game ?
Two possibilities.
Firstly, identify the dialogue by ID - as dzuigo's app does. So every time a dialogue is displayed the tool knows, 'this is file NVL_W, dialogue #4". You'd then have used some sort of a field file viewer to go through and see what dialogue it contained, and recorded a file (e.g. NVL_W_04.wav) with the dialogue for whatever dialogue #4 said.
Secondly, identify the dialogue by text - more like my test app. So when the dialogue is displayed the tool knows, 'Cloud is saying "The black-caped man?"'. It then looks to see if you've recorded a file linked to that text and if so plays it. (It could also take into account the location you were in, of course.) You'd probably strip out a certain amount of formatting/punctuation to make matching it up easier.
The first approach is arguably more 'correct', but there are advantages and disadvantages to both approaches;
-If the text changes the first method will still play the audio, the second won't (say somebody has edited the field files). It's hard to say which is better! If all that's happened is that somebody has corrected a spelling error in the field file, you probably still want the audio to play. On the other hand, if somebody has changed the text totally you probably don't want the original audio to play! The second method would let you play a different file (if you knew what the new dialogue was), so you could in theory distribute a single voice pack that still worked (and actually played something different) when certain patches to the dialogue were installed.
-In a similar way, the first method would still work if you were running a foreign version of the game (provided the dialogue IDs matched up). Of course, would you want English dialogue on a French language game?
(That raises another issue; I know in the UK release they at least renamed 'Aerith' to 'Aeris' [thank goodness]; does anyone know if they actually changed any other dialogue? I'd assume not, but clearly it wasn't just a direct unaltered release...)
-The second method has the possibility that standard lines of text, or any lines of text said in more than one place, could be recorded just once and the tool could 'automatically' reuse them between different field locations.
The disadvantage of that would be you might not want it to do that...
-
The versions of FF7 I've played here in North America are pronouncing her name as Aeris as well.
-
First one, Dziugo, I tried to run your tool to see if I could gather more info about how the game handles dialog but when I start dllloader, i get a side-by-side error. Obviously I know this isn't even alpha yet and this is to be expected but I figured you'd like to know. I use Vista and it was in the right directories.
Need more info. What's the version of your ff7.exe? If any crash occured, the error report with addresses would be useful (if you don't want to crash FF7 again, go to "Control Panel -> Administrative Tools -> Event Viewer -> Window Logs -> Apps".
-
Dziugo : I had a bit more time to look into this and the issue was that I didn't have the Visual C++ files on my computer. Downloading and installing the 2008 vcredist_x86.exe from Microsoft's website fixed the issue. Quick test worked. I'm going to try it out for a bit and report back.
edit : Tried it out some more. The way your software works, I can definitely see this project working out technically. Brilliant.
However, the application doesn't seem to handle questions well. Tried the two questions on the highwind (operations room and take control of the helm) and they both send the game into an infinite loop. Here's the start of my debug file showing a loop :
MapID = 74 DialogID = 18 Just started playing.
MapID = 74 DialogID = 18 Has been stopped.
MapID = 70 DialogID = 102 Just started playing.
MapID = 70 DialogID = 102 Has been stopped.
MapID = 70 DialogID = 114 Just started playing.
MapID = 70 DialogID = 114 Has been stopped.
MapID = 70 DialogID = 112 Just started playing.
MapID = 70 DialogID = 112 Has been stopped.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing
...
It keeps on going this way until I close the game. The actual text in game never shows up and the sound plays each second though!
Looking at the source of your program (installed VC++ Express 2008 out of curiosity to try and understand what you did), I'd wager it'd be related somehow to the program interpreting the part where ff7 waits for input as a end dialog and the program trying to reinitialize again since the dialog never came up in the first place. Only a theory from somebody who doesn't know much about these things and who's trying to understand.
To avoid that whole problem completely, would it be possible to exclude the ASK opcode messages from being picked up ? I don't remember a spot in the game where a question would be important enough to need to be voiced.
Ficedula : Wouldn't the text match way create a lot of overhead if the program has to look into a big heavy text file to match the clip before sending the sound out ? Wouldn't the ID way be more responsive ?
Ficedula & Dziugo : I've noticed, ff7music picks up the log Dziugo's hook creates in its main window (where all the texture stuff and mid info is written). With this in mind, would it be easier to use ff7music built in "sorter" and output engines in conjunction with Dziugo's hook or would it be easier to make one from scratch using Dziugo's program as a base ?
Here's a short copy & paste of that window (removed some redundant lines) :
TEXTURE LOADED INTO VIDEO MEMORY
MIDI play: 47
reading midi file: HIKU.mid
HIKU.mid
Midi file matched! HIKU
CSA6
CSR7
File counterpart 4-03 Airships Make Me Happy - The Highwind Takes to the Skies Remix.mp3
Resolved to F:\Documents\Music\ff7\4-03 Airships Make Me Happy - The Highwind Takes to the Skies Remix.mp3
Playing music
CSA6
CSR7
CSA8
Look for plugin for mp3
Init plugins
Setup input plugin: Nullsoft MPEG Audio Decoder 2.96+AAC
Setup output plugin: waveOut output v2.0.2a
Init finished
---------------------------
Prevented from resuming
CSR9
current volume: 127
MIDI is playing SUCCESSFULLY!!!
TEXTURE LOADED INTO VIDEO MEMORY
TEXTURE LOADED INTO VIDEO MEMORY
MIDI play: 47
song is already playing...
TEXTURE LOADED INTO VIDEO MEMORY
TEXTURE LOADED INTO VIDEO MEMORY
MapID = 74 DialogID = 18 Just started playing.
TEXTURE LOADED INTO VIDEO MEMORY
TEXTURE LOADED INTO VIDEO MEMORY
MapID = 74 DialogID = 18 Has been stopped.
TEXTURE LOADED INTO VIDEO MEMORY
TEXTURE LOADED INTO VIDEO MEMORY
MIDI play: 47
song is already playing...
TEXTURE LOADED INTO VIDEO MEMORY
TEXTURE LOADED INTO VIDEO MEMORY
MapID = 70 DialogID = 102 Just started playing.
MapID = 70 DialogID = 102 Has been stopped.
TEXTURE LOADED INTO VIDEO MEMORY
MapID = 70 DialogID = 114 Just started playing.
MapID = 70 DialogID = 114 Has been stopped.
MapID = 70 DialogID = 112 Just started playing.
MapID = 70 DialogID = 112 Has been stopped.
TEXTURE LOADED INTO VIDEO MEMORY
TEXTURE LOADED INTO VIDEO MEMORY
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
MapID = 70 DialogID = 1 Just started playing.
... (loop 'til alt-f4 as explained previously).
I do feel I should point out I tried the hook with and without ff7music and the loop still starts in either case so it would seem unrelated.
-
To avoid that whole problem completely, would it be possible to exclude the ASK opcode messages from being picked up ? I don't remember a spot in the game where a question would be important enough to need to be voiced.
I was thinking the same thing, but was dumb enough not to check explicitly if we're dealing with MESSAGE opcode. Now I'm curious to why it gets into that infinite loop :)
Ficedula & Dziugo : I've noticed, ff7music picks up the log Dziugo's hook creates in its main window (where all the texture stuff and mid info is written). With this in mind, would it be easier to use ff7music built in "sorter" and output engines in conjunction with Dziugo's hook or would it be easier to make one from scratch using Dziugo's program as a base ?
It picks up debug messages because I debug the same way the FF7 does (OutputDebugString). And to answer your question - that could work. And one of it's advantages is that if the app crashes for whatever reason, FF7 would still be running. I, however, was thinking about running completely in the FF7 address space. Worked out a simple format for storing files (hashtables for speeding up the file access), just don't know anything about playing sound (you can see that easily with a "PlaySound" stub in the project :P) and don't want to use external library for playing MP3s. I was looking at some open source MP3 players, but maybe we'll get input from ficedula as he's more experienced coder than I am, and most likely already got his vision on how it could be done.
-
Running out of process would be my preference - mainly so if the code does crash, it doesn't affect FF7 at all, you just lose the voicing. Obviously there'd be a small code stub in-process transmitting the relevant messages out to the main program (that's how my test app worked...) although I didn't use OutputDebugString because I've sort of gone off that as a way for IPC.
For playing sounds then to be honest for anything beyond the capabilities PlaySound gives you, I'd be tempted to use an external library just for ease of use. With that said PlaySound might actually just work, it's pretty primitive but it's not like this project needs anything too advanced...
Ficedula : Wouldn't the text match way create a lot of overhead if the program has to look into a big heavy text file to match the clip before sending the sound out ? Wouldn't the ID way be more responsive ?
Not particularly, to be honest. If you're in location NIV_W, say, then the first thing the program needs to look up is all the dialogue for that location - and any one location isn't going to have that much dialogue. It won't have to even consider dialogue for other locations, if we don't want it to.
As a further optimisation, in most cases we'd also know who was speaking (Aeris, Cloud, etc.) and we'd only have to look at dialogue for that character, too.
If you wanted to go on and search some 'general' bank of dialogue that might be applicable to all locations that might be slightly slower, but again you'd only search dialogue relevant to that specific character.
Running out of process also has the advantage that this is all happening in the background without affecting the main game too much.
-
PlaySound probably won't play MP3s, it's just there to give an impression on how it could work :) I see no problem with using a library for playing MP3s, as long as it's not an external library.
I'm looking for a compact project like:
1. Download
2. Unpack
3. Play
Not:
1. Download
2. Download xxx from www.xyz.com, move xxx.dll to the folder with exe
3. Download yyy from www.zyx.com, move yyy.dll to the folder with exe
4. etc.
X. Play
It would be a single library, but still :)
-
PlaySound can play MP3s provided they're stored in a WAV container. It'll basically play any WAV file - assuming the right codec has been installed. MP3 is of course a pretty safe codec to rely on, it's been shipping with Windows & Media Player for years...
If we needed to use a more unusual codec, then yes - ideally you wouldn't want the user to have to download anything themselves, if it could just be a DLL that sits alongside the other files we can redistribute that would be best.
-
It can? Oh, my bad.
-
So, here, have another proof of concept;
http://www.sylphds.net/f2k3/voices.rar
How to use;
Download, decompress into your FF7 folder. Run FF7. Alt-tab out (it seems to work OK with the windowed mode patch too) and run Voices. Click the 'start' button. If you don't see any errors messages then (hopefully) that means everything is working.
The window will display any dialogue it intercepts; there's some sample dialogue for the Highwind bridge I recorded in there as well (WARNING: rubbish recordings!). Easiest way to hear that would be to download this savegame:
http://www.sylphds.net/f2k3/save03.ff7
...and load the Highwind save, go to the bridge, then talk to Barret/Cid/one of the engineers.
Feedback welcomed. Preferably of the sort 'it works 100%!' but other feedback would be OK too, I guess... ;)
-
Just to clear this up, I realize it's hard to fit a british accent into the Game, but Reeve would be a good choice for a british accent. If you want to voice one of the main characters, you would have to learn more of a north american dialect :P
Oh, I don't know about that. I don't think having an accent should deter people, let's see what they've got and then we'll decide if it fits or not. They can try both! Besides, the only reason that the characters have North American accents is because the dubbing was done in the United States. If Square had their main English division in Europe, we'd be hearing something much different. Foreign accents make Gaia really seem like a different world. And who says the characters are like us? =) After all, they're originally Japanese. If those voice actors started speaking English, hehehe...
As for Aeris/Aerith, wasn't there a point in the US game when you learned that Aeris' true name was Aerith, similar to learning Red XIII was Nanaki? Their working names stayed the same, as far as I remember. Either way, both Aeris and Aerith are romanized as "e-a-ri-su," pronounced "eh-ah-ree-soo." There's no way to convey a difference between "-s" and "-th" using the Japanese alphabet, so it's mainly a stylization, whatever you choose. Something akin to writing "Katamari Damashii" as "Katamari Damacy." You could even write her name as Earth, as that's what it's supposed to represent.
Personally, I prefer the name Aeris as it rolls off the tongue better and that's how I remember it being in the original game. I realize Square changed it in all media after Final Fantasy VII, but if Square Japan had a problem with how the name was anglicized originally, you'd think they would have said something. I've heard they're quite involved in the translation progress, at least with the later games.
-
I'm sure I remember reading somewhere, that Aeris' name in the European/American release was to be Aerith, but apparantly, as you say, Aries "rolled of the tongue" better. I personally prefer Aerith, I have no idea why, but anytime I play it, I change her name to Aerith. I just think it's because it's closer to Earth, which was her name originally, or at least, supposed to be.
However, I'm sure people don't want to hear my half drunk ramblings on Aeris vs Aerith, so I'll leave it at that. :-)
-
So, here, have another proof of concept;
http://www.sylphds.net/f2k3/voices.rar
How to use;
Download, decompress into your FF7 folder. Run FF7. Alt-tab out (it seems to work OK with the windowed mode patch too) and run Voices. Click the 'start' button. If you don't see any errors messages then (hopefully) that means everything is working.
The window will display any dialogue it intercepts; there's some sample dialogue for the Highwind bridge I recorded in there as well (WARNING: rubbish recordings!). Easiest way to hear that would be to download this savegame:
http://www.sylphds.net/f2k3/save03.ff7
...and load the Highwind save, go to the bridge, then talk to Barret/Cid/one of the engineers.
Feedback welcomed. Preferably of the sort 'it works 100%!' but other feedback would be OK too, I guess... ;)
It does work a 100%. And I do see what you mean by "English Accent". ;)
Quick question though. There is perhaps a 1 second delay before the voice starts which I don't recall from dziugo's app, although that might be from a small delay in your wav files recording. I'll switch the file to one I know has no startup delay and post back.
edit : ignore this part, it was indeed the delay in the sound file.
Also, when used in conjunction with ff7music, the file doesn't play and I get an access violation at address 0000000 when it should play a file (busy soundsystem because of ff7music ?)
edit : the access violation didn't happen with the test wav I put in there (changed all dialog to it to test). I reverted back to your original files afterwards and they don't play now if ff7music is playing but they don't give errors either anymore ...
-
I'm sure I remember reading somewhere, that Aeris' name in the European/American release was to be Aerith, but apparantly, as you say, Aries "rolled of the tongue" better. I personally prefer Aerith, I have no idea why, but anytime I play it, I change her name to Aerith. I just think it's because it's closer to Earth, which was her name originally, or at least, supposed to be.
However, I'm sure people don't want to hear my half drunk ramblings on Aeris vs Aerith, so I'll leave it at that. :-)
Well, I've played and completed the USA release of Crisis Core Final Fantasy VII and all the way through they call her Aerith not Aeris and also it's easier to start the next word of a sentence in my opinion.
I don't know how good i'd be as a voice actor, i'll have a look round but my friend is interested in this project and has done voice acting before, I think she'd make a good Aerith but I don't know what she wants to do. She's quite flexible with her voice so just say the word!
-
Even if you're not a voice actor Jordie, you can still come to show your support, as the attendance we receive will determine whether or not this project moves forward. (Same statement goes for your friend too)
If you don't want to be a voice-actor, everyone will be filling out comment cards as each person auditions to help determine the cast of voice-actors.
-
I've been doing further testing with fice's proof of concept. Mostly with catching dialog since I have no mic that can make me audible for now.
It works mostly well. It actually catches dialog great although in area zz4, when vincent meets Lucrecia for the first time, it somehow didn't pick up the text but picked up the area loading. I unfortunately didn't notice it quick enough and saved over my game file afterwards so I couldn't go back to test it a second time to see if it was a fluke or a bug with the area.
Also, I've set up a test scene during the zack flashback with a test wav and I get a weird issue.
I've created a ztruck folder (name of the field file) and copied the dialogue.txt which I then modified.
However voices picks up the field file as ztruck but then changes straight away to ACKç (always the same thing) :
Changed location: ztruck
Changed location: :ACKç
The dialogue from ztruck then goes on as if nothing happened.
I'm thinking that's a game issue though as it's been perfect for every other area so far (its allowed me to identify which area is which and to start putting together a US PC game script for recording purposes).
I've also set up a test in convil_1 and that one worked great.
Great job!
PS : You can also disregard my last post regarding the "files stopped playing". I'm using my hdtv as my monitor and sometimes text can appear small. I had made a copy of the dialogue.txt file to modify it and didn't restore the name properly because I saw a "y" as an "e" from my couch! I can however say that the incompatibility between ff7music and this software is due to your wav files since my wavs works fine - I'm thinking your wavs are mp3 encoded as wavs as you mentioned earlier and since my ff7music mostly plays mp3's, I'm guessing its having an issue with playing 2 mp3 at the same time ?
-
Marcis, if you're interested in compiling a game script drop me a line. I've been working on that and am about at the Chocobo Farm right now. It's a lot of work, so if it's something you want to do, team work is going to save you many, many hours.
GlitterBerri/jack.o.lantern has also been doing some work on the finer points of it.
-
As I've mentioned, I've actually started my own while testing this one.
It starts when at the first Huge Materia quest and will go until the end of disc 3 (I'm currently at the end of the huge materia quest). I've also done the Gelnika, Cloud/Zack's flashback and Lucrecia's waterfall dialogue.
I'd be glad to send you what I've got so far if you pm me your e-mail address.
How are you guys doing this ? Are you using Cosmo or playing it ? If you're playing it I'd suggest you try Ficedula's voices as its going to both beta test it and catch all text you see on screen!
-
I'm playing through in windowed mode and simultaneously using Loveless. Loveless is great for catching the text, and it's got notes on what locations are what. Play through is just to confirm what's dummied out, order of certain things, etc. The big pain in the ass is finding what's unused, along with sorting out dialog that relies on certain events and certain people in your party. There are so many "branches", and that's what makes it poke along. For me at least, maybe you've found a faster solution.
P.S. my email is in my profile. I use IM if you want to go that route as well, and we can work out how to combine what we've got so far and how to proceed. Fortunately, Midgar might have been the slowest portion there was, and it's finished now.
-
Unfortunately, your e-mail is hidden in your profile (as in mine)!
As for loveless, I've never used it. I use ficedula's proof of concept which catches all text that gets displayed. Works pretty well too and this way I'm somewhat beta testing the software. My testing so far tells me the software works well and could even be used in its current state if needed.
As for catching text, I don't use windowed mode though. Up until this point, I've also ignored most branches and just focused on the main text. I've done a *small* portion of branching text using Cosmo but at this point I'm focusing on catching the main story and figuring out which area is which (by playing) and I plan to revisit the side-text using Cosmo.
One thing to look out for though which I've only just realized, is that Ficedula's software uses the names of the characters to sort lines. That means that some lines spoken by a character (ex: Cloud) where his name might not be used to identify the text would be listed as "none" instead of as "cloud" (name of the speaking character not written 100% of the time). The most prevalent place of this is in Cloud's mind on disc 2. If you notice it happen, we need to have the real character name beside it for voicing purposes but for technical set-up reasons, we should make a special note when that happens so we know to set it up as "none" in the program set-up.
-
you gonna join the casting call ff7heart?
-
Dont worry- I've been coming here every single day to see how things are going :D I just couldn't believe the overwhelming response! Its fantastic what everyone is doing :]
Unfortunately I dont think my voice would quite suit any of the characters-- I have an English accent, and a common one, at that xD Oh, and my North American accent is awful xDD
But I'll definitely be in Vent, filling out the comment cards and such!
-
Ficedula, how will voice-overs be programmed into some of the battle sequences? Be it at the beginning or during the victory fanfare. Some of the battles will need to have voice-overs, and if we're throwing voice-overs into the battle, I'm thinking we could implement voice-overs at the beginning of specific battles, or when a Boss begins to cast a specifc, powerful spell, or randomly in occasional battles during specific victory fanfares.
-
Ficedula, how will voice-overs be programmed into some of the battle sequences? Be it at the beginning or during the victory fanfare. Some of the battles will need to have voice-overs, and if we're throwing voice-overs into the battle, I'm thinking we could implement voice-overs at the beginning of specific battles, or when a Boss begins to cast a specifc, powerful spell, or randomly in occasional battles during specific victory fanfares.
Unsure. All of those things are probably possible to some extent, given enough time to hook the right parts of the engine.
Best way to proceed is probably simply to deal with each particular requirement as it occurs ... when you find a battle that needs voicing, get a savegame as close to the battle as possible, let me know what you'd like to happen (ideally) and what you could use (in the worst case) and I'll investigate what can be done next time I get a moment...
-
The good news is, there's dialog in the very first boss battle with the guard scorpion, so getting a save should be quick and easy.
-
Didnt have the time to read the whole thread and havent got that much time now so:
Are you still looking for some actors? Cause I badly wanna take over the voice of Sephiroth.... :wink: Since my person is dark and creepy(after years of black-metal^^) I think I would manage to do his part......
email: [email protected]
msn: [email protected](Nickname is.....big suprise.....Sephiroth(GF-Terrabreak), cause the last is my Unreal Tournament online name^^)
-
Sephiroth is taken, wouldn't mind hearing your VIncent though if you can jump on Vent. The vent info is in the Pre-Release Discussion Thread.
-
Can I try Vincent? :-D
[This place is dead and I want to ressurect it! :wink:]
-
WHO IS STILL AVAILABLE FOR AUDITIONS?
Godo and Shera. Many Characters also have one-liners who need to be voiced too.
Maybe you should do a search before your ressurections.
-
Or better yet ask in the FF7 Voice forum.
-
To tell the truth, I'd rather not have any voices in FF7.
They'd kind of sound awkward at the perspective you see the game from.
-
whoever is doing barrets voice better be black or have mastered the black language :-D :mrgreen:
-
To tell the truth, I'd rather not have any voices in FF7.
They'd kind of sound awkward at the perspective you see the game from.
Nobody is pointing a gun in your face and telling you to use the patch when it comes out.
This sort of thing is exactly why I'm freaking out about dziugo's program getting finished. Almost all of the material for Midgar is ready and it would be relatively easy for me to put together a playable demo so people could see for themselves if this is a project they'll be interested in. It's a lot better than trying to assure people the voices fit pretty well and the actors fit their roles.
And if anyone is still interested in the project, don't be afraid to ask for a role anyways. I guess a couple people flaked out and had to be replaced, so you never know.
-
so it was just a joke
-
so it was just a joke
Look at the quote. You weren't in it.
-
This sort of thing is exactly why I'm freaking out about dziugo's program getting finished. Almost all of the material for Midgar is ready and it would be relatively easy for me to put together a playable demo so people could see for themselves if this is a project they'll be interested in. It's a lot better than trying to assure people the voices fit pretty well and the actors fit their roles.
And if anyone is still interested in the project, don't be afraid to ask for a role anyways. I guess a couple people flaked out and had to be replaced, so you never know.
I think short movies should be sufficient to showcase the voices for myself. And with Dziugo's program, there won't be an actual setup per say as it uses file names. That's why area and line numbering info is needed in scripts from now on. Easier for us!
Also, I'm not aware of anybody that jumped shipped yet although we do have some minor roles left to be fillled (ex: godo, tifa's father, shera, important townspeople such as the one who takes sephiroth's picture in nibelheim, etc.). If you wish to apply, just post in the ff7voice forum.
-
I reapeat. Can I try out for Vincent :?
-
Vincent is not in my list and is not a minor caracther : he's filled.
You're more than welcome to try for any of the roles I just mentionned however.
-
Do it anyways, just in case. If nothing else, they might find your voice suitable for something else. Tifa has been a hard role to fill, I think their second person has already left. If you know anyone that might be interested in that role let them know. Doesn't hurt to be redundant.
-
I think Opine has agreed to take over Tifa after the first two scenes were completed and has been doing her lines for a few months now if I'm not mistaken.