I'm trying to make a cutscene where the characters are talking (with expression and lip sync) but with no voice and subtitles appearing. Does anyone know of a smart way to do this?
Bonus question: Are there any tutorials for FXA? I need to create a few subtle expressions between the "talking".
Having actors talk without subtitles and voice
Débuté par
Guest_dewkl_*
, juin 02 2010 11:37
Retour en haut
#2
Posté 02 juin 2010 - 01:02
Kilrogg_
-
- Members
- 296 messages
If you don't generate VO for the line you're using, does it still lip-sync? If not, then maybe you can try generating VO and deleting the generated VO file from your toolsetexport folder and see if the lip-sync is still there.
#3
Guest_dewkl_*
Posté 02 juin 2010 - 01:20
Guest_dewkl_*
Guest_dewkl_*
-
- Guests
I tried that, I'm pretty sure it requires VO to lipsync and make fxa. With your suggestion, I think the subtitles would still be there.Kilrogg_ wrote...
If you don't generate VO for the line you're using, does it still lip-sync? If not, then maybe you can try generating VO and deleting the generated VO file from your toolsetexport folder and see if the lip-sync is still there.
#4
Posté 02 juin 2010 - 01:23
BillHoyt
-
- Members
- 109 messages
You don't need the subtitles if you use .WAV voiceover, so you can just leave the conversation tree blank and there won't be anything to display on the screen. As for turning off the voice, I'm not so sure about that one... you will need some voice in order to generate the lip-sync - otherwise there's nothing to sync to.
#5
Guest_dewkl_*
Posté 02 juin 2010 - 01:35
Guest_dewkl_*
Guest_dewkl_*
-
- Guests
I'm not sure I understood your first part. How do you mean .wav voiceover? From what I can see, I would suppose it's easier to just create a line and generate VO and lipsync like normal, then delete the actual audio somehow (or replace it with nothing). As for the subtitles...I don't know.
#6
Posté 02 juin 2010 - 03:31
Beerfish
-
- Members
- 23 830 messages
A bizarre little twist to this type of thing from a post I made a short while ago. I had created fae fx and robo voice over for a cutscene. I then wanted to test recorded voice over which I did myself and which worked.
I then decided I did not want my voice over but still wanted the lip sync.
I moved the wav files out of the toolsetexport folder thinking that would do it. The voice over remained. I changed the names of the voice over files so they would not = convo string #'s the voice over remained. I deleted the voice over files. The voices remained.
After each one of these steps I redid the voice over and face fx in the toolset and did a reexport. I finally ended up recording new wav files with the proper id numbers but I did not voice them at all. I ended up with the lip sync but no voice.
I found it totally bizarre that once generated and recorded I could not get rid of my voice overs.
I then decided I did not want my voice over but still wanted the lip sync.
I moved the wav files out of the toolsetexport folder thinking that would do it. The voice over remained. I changed the names of the voice over files so they would not = convo string #'s the voice over remained. I deleted the voice over files. The voices remained.
After each one of these steps I redid the voice over and face fx in the toolset and did a reexport. I finally ended up recording new wav files with the proper id numbers but I did not voice them at all. I ended up with the lip sync but no voice.
I found it totally bizarre that once generated and recorded I could not get rid of my voice overs.
#7
Posté 02 juin 2010 - 04:41
BillHoyt
-
- Members
- 109 messages
Sorry, I should have been more clear. Say you want the guy to mouth the words "What is the capital of Assyria?"
1. Make a .WAV file of yourself asking the question (Audacity makes this very easy). It doesn't matter what it sounds like because no one will hear it anyway.
2. Give the file the number of the conversation line from the bottom left of your screen when you click on it, so if the line is "1234567890," your wave file will be "1234567890_m.wav." You don't need to have anything typed in the conversation line.
3. Put it in your module export directory.
4. Generate VO and FaceFX. Then once you figure out how to turn the voice off, the FaceFX should remain.
The problem I think with generating the computer voice and then changing the lines is that you may lose the VO/FaceFX you already generated (since the timestamp on the line is later than the VO based on the line). I don't know that for sure, but I think I did that once...
Beerfish - there is an intermediate set of export directories where a lot of that kind of stuff is hidden. It makes it almost impossible to get rid of stuff sometimes...
1. Make a .WAV file of yourself asking the question (Audacity makes this very easy). It doesn't matter what it sounds like because no one will hear it anyway.
2. Give the file the number of the conversation line from the bottom left of your screen when you click on it, so if the line is "1234567890," your wave file will be "1234567890_m.wav." You don't need to have anything typed in the conversation line.
3. Put it in your module export directory.
4. Generate VO and FaceFX. Then once you figure out how to turn the voice off, the FaceFX should remain.
The problem I think with generating the computer voice and then changing the lines is that you may lose the VO/FaceFX you already generated (since the timestamp on the line is later than the VO based on the line). I don't know that for sure, but I think I did that once...
Beerfish - there is an intermediate set of export directories where a lot of that kind of stuff is hidden. It makes it almost impossible to get rid of stuff sometimes...
#8
Posté 02 juin 2010 - 05:09
Kilrogg_
-
- Members
- 296 messages
The Emissary mod actually has a conversation at the beginning where a human talks but since you're a hurlock it comes out as silence (since you dont understand the language). Might be good to ask the creator how he did it.
#9
Posté 08 juin 2010 - 09:01
Cuvieronius
-
- Members
- 91 messages
okay, a several step method for what you want... but it works. Oddly I just did this today as a test... and it wasnt what I wanted, but is what you want it seems.
generate vo (robo) for the conversation then generate facefx. Go get the face fxe and fsm files from the override... delete the fsb and fev it generated.
Go to C:\\Users\\owner\\AppData\\Local\\Temp\\DragonAgePlaceholderVO\\en-us and copy out the conversation vo folder. Replace the robo vo files with 1 second blank wav files.
Open the FDP in that folder with FMOD and build it. It will use the blank sounds, and leave the lipsync working. You wont have any conversation text though, but sounds like that is what you want.
Put those in your override/sound folder and you are finished.
generate vo (robo) for the conversation then generate facefx. Go get the face fxe and fsm files from the override... delete the fsb and fev it generated.
Go to C:\\Users\\owner\\AppData\\Local\\Temp\\DragonAgePlaceholderVO\\en-us and copy out the conversation vo folder. Replace the robo vo files with 1 second blank wav files.
Open the FDP in that folder with FMOD and build it. It will use the blank sounds, and leave the lipsync working. You wont have any conversation text though, but sounds like that is what you want.
Put those in your override/sound folder and you are finished.
#10
Posté 08 juin 2010 - 09:47
tmp7704
-
- Members
- 11 156 messages
As a shortcut, there's also a simple "fake lipsync" animation which can be applied to character directly. It's named "cs_f.lpsync".
