A Community discussion forum for Halo Custom Edition, Halo 2 Vista, Portal and Halo Machinima
|
Home
Search
Register
Login
Member List
Recent Posts| »Forums Index »Halo Custom Edition (Bungie/Gearbox) »Halo CE General Discussion »scripting a biped to talk | 
|
| Author | Topic: scripting a biped to talk (5 messages, Page 1 of 1) | ||||
| Moderators: Dennis | |||||
 Jobalisk Joined: Feb 8, 2014 
The haiku master。 |
Posted: Dec 10, 2015 10:11 PM
Msg. 1 of 5
anyone know how to get a biped to open its mouth a few times so that it looks like its talking?
|
||||
|
|||||
|
MatthewDratt Joined: Sep 11, 2010 
TAKEDOWN IS OUT MattDratt.com |
Posted: Dec 10, 2015 10:13 PM
Msg. 2 of 5
1) the mouth needs to be rigged. If you think it might not be, just throw the sound on a h1 marine or something
2) the audio needs to have mouth data --but HOW?? Well to get mouth data is simple. Compile the sound using tool as normal. Don't move anything around. Open the sound file in guerilla, and next to "class" select something that is dialog related. Unit Dialog or Scripted Dialog should both work. Save. Recompile the sound with tool again. Reopen up the file and at the bottom you should see that mouth data has bytes. If it doesn't, you messed up. Now just script as normal (sound_impulse_start takedown\levels\extermination\dialog\alpha\a_nova_02 nova 1) Nova is the biped's name Edited by MatthewDratt on Dec 10, 2015 at 10:14 PM |
||||
|
|
|||||
|
Jobalisk Joined: Feb 8, 2014
The haiku master。 |
Posted: Dec 10, 2015 10:15 PM
Msg. 3 of 5
Quote: --- Original message by: MatthewDratt 1) the mouth needs to be rigged. If you think it might not be, just throw the sound on a h1 marine or something 2) the audio needs to have mouth data --but HOW?? Well to get mouth data is simple. Compile the sound using tool as normal. Don't move anything around. Open the sound file in guerilla, and next to "class" select something that is dialog related. Unit Dialog or Scripted Dialog should both work. Save. Recompile the sound with tool again. Reopen up the file and at the bottom you should see that mouth data has bytes. If it doesn't, you messed up. Now just script as normal (sound_impulse_start takedown\levels\extermination\dialog\alpha\a_nova_02 nova 1) Nova is the biped's name Edited by MatthewDratt on Dec 10, 2015 at 10:14 PM The model in question is a standard marine, so hopefully this will work, thanks and it did, thanks! Edited by Jobalisk on Dec 10, 2015 at 10:35 PM |
||||
|
|
|||||
|
Kinnet Joined: Dec 27, 2013 
why are we still here? just to suffer? |
Posted: Dec 11, 2015 10:02 AM
Msg. 4 of 5
Quote: --- Original message by: sparky Mouth data is just a 30-per-second processing of average amplitude (volume) of the sound. It divides each second of the sound into 30 sections and looks at each section's average amplitude, then stores the section's amplitude as a ratio of 0-100% within 1 byte as 0-255, so 0% amplitude (silence) is stored in a byte as 0, and 100% amplitude (maximum volume) is stored in a byte as 255, so you have 30 bytes of amplitude values per second of sound sample duration. That's the mouth data: the simple volume of the sound translates to how "open" the mouth is, so that louder sounds translate into larger mouth openings. I was working on an experimental version of Zeus which would let you modify the mouth data. I even have a video of it. By modifying the mouth data directly, you can have something like a jaw hanging open in silence -- or you could just import the sound as previously explained, letting Tool process it, then change it to another usage class type, like "music", and import a modified sound over it, giving you a modified sound with the processed sound data; I don't know if Tool would erase the mouth data if you re-imported a sound as a non-dialogue class. How many books do you read daily ? |
||||
|
|
|||||
|
Skidrow925 Joined: Mar 19, 2010 
"ideological sense of respect and tact of a 5yo" |
Posted: Dec 13, 2015 02:23 PM
Msg. 5 of 5
Quote: --- Original message by: Kinnet Quote: --- Original message by: sparky Mouth data is just a 30-per-second processing of average amplitude (volume) of the sound. It divides each second of the sound into 30 sections and looks at each section's average amplitude, then stores the section's amplitude as a ratio of 0-100% within 1 byte as 0-255, so 0% amplitude (silence) is stored in a byte as 0, and 100% amplitude (maximum volume) is stored in a byte as 255, so you have 30 bytes of amplitude values per second of sound sample duration. That's the mouth data: the simple volume of the sound translates to how "open" the mouth is, so that louder sounds translate into larger mouth openings. I was working on an experimental version of Zeus which would let you modify the mouth data. I even have a video of it. By modifying the mouth data directly, you can have something like a jaw hanging open in silence -- or you could just import the sound as previously explained, letting Tool process it, then change it to another usage class type, like "music", and import a modified sound over it, giving you a modified sound with the processed sound data; I don't know if Tool would erase the mouth data if you re-imported a sound as a non-dialogue class. How many books do you write daily ? |
||||
|
|
|||||
|
