Jump to content

DoubleCheeseburger's xVASynth Based Mod Voicification Thread


Recommended Posts

3 hours ago, DoubleCheeseburger said:

Updated BaboDialogue voice files based on version 4.031. The link is in the first post.

It's actually surprisingly 2 Gb smaller now, thanks to the new version of the export script, which I will be releasing soon.

The voice files for babo 3.901 had two files, one for generic and one for unique dialogue files. Does 4.031 encompass both of those previous files?

Link to comment
8 hours ago, applesandmayo said:

The voice files for babo 3.901 had two files, one for generic and one for unique dialogue files. Does 4.031 encompass both of those previous files?

Yes, it does. But I stopped generating voice files for the most of the unique voices unless a dialogue line directly references it.

For example, there are only 15 Ancano voice files because those lines directly refer to that NPC. Other ~2.5k of "generic" dialogues lines are not generated for Ancano.

Link to comment
1 hour ago, m5holmez said:

Sorry if I am sounding like a halfwit here but for some of these mods listed I edited a lot of the dialogue, I assume this will cause an issue with these voiced lines? 

The only one I've seen mess with the vanilla voice lines is the cutting room floor voice pack. Its most noticeable when you see Uthgerd talking to that horse dude about Cicero.

Link to comment
16 hours ago, applesandmayo said:

The only one I've seen mess with the vanilla voice lines is the cutting room floor voice pack. Its most noticeable when you see Uthgerd talking to that horse dude about Cicero.

 

No what I mean was that if I have personally edited the dialogue lines in some of these mods will that mess with these voiced lines, I don't know how the .fuz files work do they just read the dialogue lines in the ESP? 

Edited by m5holmez
Link to comment
1 hour ago, m5holmez said:

 

No what I mean was that if I have personally edited the dialogue lines in some of these mods will that mess with these voiced lines, I don't know how the .fuz files work do they just read the dialogue lines in the ESP? 

 

The .fuz files are a merge of lip-syncing(face mesh morphs) and audio files, likely xWMA(DirectX audio).   

The fuz file name match the dialogue formid path and is located inside sound/voice/mod plugin name/NPC voice name.  

 

During gameplay the dialogues will show the overridden (your edition) dialogue in subtitles (if enabled) but play the already encoded lip sync and audio. (Play original audio).

 

Game will not break it just the audio and the subtitle will not match. 

 

Edited by safado
Link to comment
2 hours ago, safado said:

 

The .fuz files are a merge of lip-syncing(face mesh morphs) and audio files, likely xWMA(DirectX audio).   

The fuz file name match the dialogue formid path and is located inside sound/voice/mod plugin name/NPC voice name.  

 

During gameplay the dialogues will show the overridden (your edition) dialogue in subtitles (if enabled) but play the already encoded lip sync and audio. (Play original audio).

 

Game will not break it just the audio and the subtitle will not match. 

 

 

Damn that's a shame as a lot of these mods listed have poorly written dialogue and bad grammar. I edited the Baka mods for example in xEdit (only changing the dialogue lines not formid's) to correct this but that will suck if these voice lines just play the original unedited lines of said mods. 

 

Would this still be the same issue if the formid's aren't changed just the dialogue lines?

Edited by m5holmez
Link to comment
6 hours ago, DoubleCheeseburger said:

Yes, the audio is standalone from the text, so it will stay the same no matter what you do to text.

 

Ah ok, so how is this done exactly if it's not reading from the dialogue lines in the .ESP of said mod? Is there any other way to say convert the .fuz files and edit the dialogue the audio is using? Obviously I am missing something here just trying to understand. 

 

Thanks. 

 

(Edit - Nevermind, had to look into it, seems it pulls all the dialogue from xEdit then compiles it so will need to re-do the whole thing with my edited dialogue in VASynth, hopefully I can figure it out) 

Edited by m5holmez
Link to comment

I've been successfully using the export script (thanks a lot by the way) but I've noticed there are some edge-cases where I think modified/added lines from existing vanilla dialogue topics won't play because I assume they have to go under the original folder (like skyrim.esm) instead. Is there a convenient way of isolating these?

Link to comment
1 hour ago, TX3 said:

I've been successfully using the export script (thanks a lot by the way) but I've noticed there are some edge-cases where I think modified/added lines from existing vanilla dialogue topics won't play because I assume they have to go under the original folder (like skyrim.esm) instead. Is there a convenient way of isolating these?

The updated version of the script should handle this.

I will be uploading it this thursday probably.

Link to comment
7 hours ago, DoubleCheeseburger said:

The updated version of the script should handle this.

I will be uploading it this thursday probably.

Thank you for the awesome script and tutorial! Just one other thing I noticed with the script (apologies if this has already been brought up). It seems that putting a "..." by itself in a dialogue line will make it so the lines that come after it aren't voiced. I noticed this in Dovakhiin's Infamy when it happened with two different characters. I'm guessing that the script sees the ellipses as a blank line and assumes that's the end of the dialogue.

 

I also found the rare mod that uses curly braces, so I added this to the filters:

  filterRegExps.Add('{.*?}');

 

Link to comment
On 11/3/2021 at 6:20 AM, throwmeinthetrash6 said:

Just one other thing I noticed with the script (apologies if this has already been brought up). It seems that putting a "..." by itself in a dialogue line will make it so the lines that come after it aren't voiced. I noticed this in Dovakhiin's Infamy when it happened with two different characters. I'm guessing that the script sees the ellipses as a blank line and assumes that's the end of the dialogue.

Hmm, I'm unaware of this, is there some specific dialogue line you're having problem with, so I could check it myself?

Link to comment

Updated SexLab Solutions Revisited voice files. The download link is in the first post.

Updated SexLab SexLab Confabulation voice files. The download link is in the first post.

 

Updated xEdit export script. Changes are:

  1. IMPORTANT! The script will now automatically create all the necessary directories for you. You no longer need to create folders manually. Also, xVASynth output path doesn't matter now. But for all of this to work you need to specify your desired output directory by editing the audioOutputPath variable in the script.
  2. Better dialogue filtering. The script is now able to analyze scenes and extract speakers from them. Also, the script is now able to extract speakers from "ANAM - Speaker" attribute. Pointed out by @Mister X.
  3. Better master record handling. "Dirty" edits or edits that don't change the actual text will no longer be exported even if those edits were made based on Unofficial Skyrim Patch. There are trustedVoicedPlugins variable now with the list of voiced plugins, but default value (which contains official DLCs and USSEP) should work fine almost all the time for Skyrim SE mods.
  4. There is now a "low priority" voices list. These are voices that will be ignored by script unless a dialogue line directly refers to voices from this list. This prevents voice packs from becoming too large. See lowPriorityVoices.CommaText. Suggested by @Mister X.

The download link is in the guide post.

 

Updated the guide itself to reflect changes made to the script.

Removed the .wav preprocessing step using SynthUtil from the guide. It's either some kind of insane luck or this step is no longer required when using xVASynth 2.0.0+.

Basically, the whole process should be much easier now once you get a grip on it.

 

Link to comment
9 hours ago, DoubleCheeseburger said:

Hmm, I'm unaware of this, is there some specific dialogue line you're having problem with, so I could check it myself?

 

So far I've only noticed this issue with the mod "Dovahkiin's Infamy." With that mod installed (and the voice pack from Executaball), if you travel to Riverwood and go near the Sleeping Giant Inn, you should trigger a quest called "The Blacksmith and the Drunkard." If you speak to Sigrid after the intro scene occurs and say "Does Embry always bother you?" followed by "I can try to convince him to leave you alone," you should see the issue I'm talking about. She'll start her response with "..." and then the following two lines won't be voiced.

 

I'm not sure if this is an issue with your script or not as that voice pack didn't come from you, but I figured that Executaball was also using your script.

Link to comment

I have issues with the BaboDialogue pack (repackaged as BSA for LE). For some situations it works perfectly fine but for others, characters basically just stand around when there should be dialogue playing. One example is the Winterhold event with the Argonians. Could that be a result of those lines missing?

 

Removing the voice pack makes the mod work fine so I know it's not the mod itself. It's certainly possible that I'm not creating the BSA file correctly using bsarch but it seems I was successful in doing so for mods with smaller voice packs.

 

Regardless, thanks for the great work so far!

Edited by krafs
Link to comment
8 hours ago, krafs said:

I have issues with the BaboDialogue pack (repackaged as BSA for LE). For some situations it works perfectly fine but for others, characters basically just stand around when there should be dialogue playing. One example is the Winterhold event with the Argonians. Could that be a result of those lines missing?

 

Removing the voice pack makes the mod work fine so I know it's not the mod itself. It's certainly possible that I'm not creating the BSA file correctly using bsarch but it seems I was successful in doing so for mods with smaller voice packs.

 

Regardless, thanks for the great work so far!

 

Do not know what triggering your issue but,

It your BSA bigger than 1.97GB,  it should reach 3.8GB ?  I split mine in two 1.97GB and that scene plays. (The one with nords next to dead argonians) 

It your BSA loaded with a dummy plugin, or it is named like the main mod?  I'm using two dummy plug ins to load the BSA into the Gamebrio engine virtual file system.

 

I use SkyrimSE. 

 

 

Link to comment
On 11/4/2021 at 7:58 PM, throwmeinthetrash6 said:

 

So far I've only noticed this issue with the mod "Dovahkiin's Infamy." With that mod installed (and the voice pack from Executaball), if you travel to Riverwood and go near the Sleeping Giant Inn, you should trigger a quest called "The Blacksmith and the Drunkard." If you speak to Sigrid after the intro scene occurs and say "Does Embry always bother you?" followed by "I can try to convince him to leave you alone," you should see the issue I'm talking about. She'll start her response with "..." and then the following two lines won't be voiced.

 

I'm not sure if this is an issue with your script or not as that voice pack didn't come from you, but I figured that Executaball was also using your script.

Wow, that's quite some outstanding observation you got there.

This was actually the case.

I generated lines that were missing due to this bug for Dovahkiin's Infamy. Attached to this post. That's just 14 lines, just add it next to the main package.

 

The script was breaking the loop instead of skipping current iteration when handling an inaudible string. Should be pretty minor. I don't see many inadudiable strings and inaudiable strings followed by audible ones should be even more rare.

Added the fixed version of the script to the guide post.

Dovahkiins Infamy v2.0 Missing Voice Files.zip

Edited by DoubleCheeseburger
Link to comment

Added voice files for highly requested Relationship Dialogue System.

The download link is in the first post.

It's pretty huge, which is to be expected, since this is pretty much a sandbox mod.

I remember building relationships with unique NPCs being pretty fun, so I generated voice files for every single voice available right now, excluding dragons.

Whole package is split into two archives: one is for generic voices and another one is for unique voices.

 

Also, this is the first package made using .lip and .fuz plugin for xVASynth v2 for lip sync files. I tested it ingame briefly and it seemed to working fine, so I guess I'll be using it on a permanent basis. Will update the guide once again when I'll have some time.

Link to comment
2 hours ago, DoubleCheeseburger said:

Wow, that's quite some outstanding observation you got there.

This was actually the case.

I generated lines that were missing due to this bug for Dovahkiin's Infamy. Attached to this post. That's just 14 lines, just add it next to the main package.

 

The script was breaking the loop instead of skipping current iteration when handling an inaudible string. Should be pretty minor. I don't see many inadudiable strings and inaudiable strings followed by audible ones should be even more rare.

Added the fixed version of the script to the guide post.

Dovahkiins Infamy v2.0 Missing Voice Files.zip 315.26 kB · 0 downloads

 

Awesome, glad I could help. And thanks for uploading those missing lines!

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   1 member

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. For more information, see our Privacy Policy & Terms of Use