DoubleCheeseburger's xVASynth Based Mod Voicification Thread

[stas2503]

On 6/20/2021 at 3:54 AM, DoubleCheeseburger said:

How to Make Voice Files Yourself

Instead of using SK, I installed the lip_fuz plugin.
XVASynth already generates .fuz files right away

So I just needed xEdit and your script to generate the csv.

Well, of course, XVA for generating voice files.

Thanks for the instructions. It was very useful.

[stas2503]

8 hours ago, DarkRavage said:

Hi Cheeseburger, can  you please add voices to the mod Reward Your Follower? This mod is extremely underrated. The SSE version is in the comments at last couple pages. 

 

Try it. Made it as a test.

RewardYourFollowers.bsa

[wower2345]

How would I go about extracting some simple follower lines with this? Extracting the entirety of Skyrim.esm seems like a chore + I would have to separate the lines somehow. I just want things like combat, world comments, and greetings.

Edited February 14, 2024 by wower2345

[DarkRavage]

9 hours ago, wower2345 said:

How would I go about extracting some simple follower lines with this? Extracting the entirety of Skyrim.esm seems like a chore + I would have to separate the lines somehow. I just want things like combat, world comments, and greetings.

Just did a quick test and it works great! Thanks for the quick response! You should post this somewhere so others may be aware of it.

[kamithemoon]

On 2/14/2024 at 1:03 PM, wower2345 said:

How would I go about extracting some simple follower lines with this? Extracting the entirety of Skyrim.esm seems like a chore + I would have to separate the lines somehow. I just want things like combat, world comments, and greetings.

Use Lazy voice finder to see and extract individual lines without having to go through the entire bsa.

 

https://www.nexusmods.com/skyrimspecialedition/mods/8619

[nilead]

Can anyone share the Xedit script linked in the guide post? Link seem to have expired.

[51offthescale]

On 5/27/2024 at 1:10 PM, nilead said:

Can anyone share the Xedit script linked in the guide post? Link seem to have expired.

 

I can still download the script in the post.

[51offthescale]

I updated SLSF comments to 3.60. This is all lines not just the difference between 3.60 and the version in the first post. The files have been packed into BSAs.

https://drive.google.com/file/d/1yZKvcSLfNEQgylwl5WejQ2kqnY-nXpUV/view?usp=drive_link

[nilead]

13 hours ago, 51offthescale said:

I can still download the script in the post.

My appology, i tried to do so using someone quoting it and it failed for some reason. Using the link directly in the post worked as intended.

[Zer0all]

Hi, i just want to say thank you a lot for all the voice pack, make all mods more fun.

Just a question tho, does sexlab eager NPCS voice pack miss some voices likes it work fine for me for npcs and all but when it come to follower there is no voice, exemple: when with slen i asked we should rest here, follower say alright, you're the boss and there is no sound. just to be sure is it on my end or the voice pack miss some files ?

maybe unique follower are not recognise in the voice game files like a normal npcs and dont work ?

 

Edited May 30, 2024 by Zer0all

[Seeker999]

2 hours ago, Zer0all said:

Hi, i just want to say thank you a lot for all the voice pack, make all mods more fun.

Just a question tho, does sexlab eager NPCS voice pack miss some voices likes it work fine for me for npcs and all but when it come to follower there is no voice, exemple: when with slen i asked we should rest here, follower say alright, you're the boss and there is no sound. just to be sure is it on my end or the voice pack miss some files ?

maybe unique follower are not recognise in the voice game files like a normal npcs and dont work ?

 

Not all voices in Skyrim are equal when it comes to making voice packs (speaking as a user, not a creator). Some vanilla npc's don't voice lines in some of my mods, while most of them do. Which is one reason you need Fuz Ro D-oh. Your follower might fall in this category.

If by unique follower you mean a modded follower that doesn't use a vanilla Skyrim voice type then they won't speak the lines from the mod.

[DuskWanderer]

Any chance for an update/patch for SL Fame Comments ver. 3.6? The last covered version was 3.31, so a patch to address everything added since then would be nice. Considering the file is already massive however I can see it being a very low priority.

[IdosheJungo]

6 hours ago, DuskWanderer said:

Any chance for an update/patch for SL Fame Comments ver. 3.6? The last covered version was 3.31, so a patch to address everything added since then would be nice. Considering the file is already massive however I can see it being a very low priority.

I believe that's already been done 

 

[DuskWanderer]

1 hour ago, IdosheJungo said:

I believe that's already been done 

 

That links to this very thread, which like I said only has version 3.31 of Comments, not the current version of the mod which is 3.6. A lot of lines have been added since then that are currently unvoiced. Most annoyingly is the Bandit/Forsworn combat lines that are supposed to play often.

Edited July 26, 2024 by DuskWanderer

[IdosheJungo]

12 hours ago, DuskWanderer said:

That links to this very thread, which like I said only has version 3.31 of Comments, not the current version of the mod which is 3.6. A lot of lines have been added since then that are currently unvoiced. Most annoyingly is the Bandit/Forsworn combat lines that are supposed to play often.

I think the links underneath the one linking to this thread takes you to this post 

 

 

[Gudulba]

Do you guys enable "Use SR" and "Use Clean-up" in xVASynth when generating voices?

 

Enabling them seems to double the processing time but I couldn't make out any difference in sound quality...

 

[Gudulba]

And here's another one... For testing purposes, I have run the voice-generating for Treasure Hunter Whore. I used Mister X's script (https://www.loverslab.com/topic/168547-doublecheeseburgers-xvasynth-based-mod-voicification-thread/?do=findComment&comment=3840008) to extract the dialogue topics and then ran it through xVASynth.

 

I then compared my output to the already available voice pack (https://www.loverslab.com/files/file/23827-generated-voice-pack-for-treasure-hunter-whore/). In-game, both voice packs seem to sound similar after a very quick test.

 

However my voice pack is about three times bigger than the downloaded one. In my pack, there are a lot more fuz files. Did I do something wrong or is the downloaded voice pack missing lines or voices?

 

Here are both voice packs if anyone would like to compare them: https://we.tl/t-UzzITvSECm

Edited August 2, 2024 by Gudulba

[petronius]

9 hours ago, Gudulba said:

However my voice pack is about three times bigger than the downloaded one. In my pack, there are a lot more fuz files. Did I do something wrong or is the downloaded voice pack missing lines or voices?

 

Without comparing, I can say that the script uses a catch-all method to ensure that you don't have any silent lines, that means generating voiced lines that will never be heard. The script does not discriminate, for each line, which voices are actually able to use it - it makes each line be generated for every voice type that's listed in the script. This is useful for mods like radiant prostitution or approach-type mods where you can have dialogue with any NPC, but if you have a mod whose lines are spoken only by a few actors (say, only guards, and you know only a few voices are used by guards) you can edit and save the script before generating in xEdit, making sure only the voices you want are used, cutting down generation time and file bulk. If the mod's dialogue is really meant to be used by only one actor, and you have only a hundred or lines to generate, it might be worth generating each line, trying to get it more life-like with arpabet and even using audacity to improve the quality?

[Gudulba]

15 hours ago, petronius said:

 

Without comparing, I can say that the script uses a catch-all method to ensure that you don't have any silent lines, that means generating voiced lines that will never be heard. The script does not discriminate, for each line, which voices are actually able to use it - it makes each line be generated for every voice type that's listed in the script. This is useful for mods like radiant prostitution or approach-type mods where you can have dialogue with any NPC, but if you have a mod whose lines are spoken only by a few actors (say, only guards, and you know only a few voices are used by guards) you can edit and save the script before generating in xEdit, making sure only the voices you want are used, cutting down generation time and file bulk. If the mod's dialogue is really meant to be used by only one actor, and you have only a hundred or lines to generate, it might be worth generating each line, trying to get it more life-like with arpabet and even using audacity to improve the quality?

Thank you for this detailed input! Is there a way to find out which voices are really used in a mod?

[petronius]

6 hours ago, Gudulba said:

Thank you for this detailed input! Is there a way to find out which voices are really used in a mod?

You can usually tell by conditions in the quest that implements the PC dialogue, or NPC idle banter lines, which you can see with both the CK and xEdit. If none are present, then it's likely that the line can be used by any NPC. If there are conditions, check what they limit those lines to: a character, a faction, etc.

 

There may be other scripts around that automate this, but I haven't found or used them.

 

Personally, though I'm all for quality when I can, I just add so many mods to my build that I don't have time to invest in them, so I download voice support when existing, or generate voice files using the exact same script you use. I make sure to always pack fuz files into bsa archives, even if you don't pack anything else, if you run a lot of xVASynth voiced mods using this method, it might save you having a million tiny loose files being parsed every time your mod manager refreshes - it can bog it down to a halt if you don't.

[applesandmayo]

Is there an updated version of the Babo/Baka mods? I'd love to see BaboDialogues voice pack updated to 3.47.2, SLApproach as well.

[IdosheJungo]

9 hours ago, applesandmayo said:

Is there an updated version of the Babo/Baka mods? I'd love to see BaboDialogues voice pack updated to 3.47.2, SLApproach as well.

Not yet, but I'm working on babo/slap atm, maybe out in a couple of days.

[IdosheJungo]

On 8/3/2024 at 6:05 PM, applesandmayo said:

Is there an updated version of the Babo/Baka mods? I'd love to see BaboDialogues voice pack updated to 3.47.2, SLApproach as well.

SLAP is up now

[applesandmayo]

14 hours ago, IdosheJungo said:

SLAP is up now

Hell yeah, awesome! Thank you!

[51offthescale]

On 8/24/2022 at 7:14 AM, kevindevon said:

Hi all i was extracting dialogue for SLSF which has alot of comments and i got this error in the script and it never finished:

 

 Exception in unit  line -1: Error in unit 'xVASynthDialogueExport' on line 132 : Undeclared Identifier 'Count'

Error during Applying script "xVASynth Dialogue Export": Error in unit 'xVASynthDialogueExport' on line 132 : Undeclared Identifier 'Count'

 

is there something i need to change in the script ?

 

any help at all be much appreciated

 

Thanks 

 

Hey did you ever figure this out? I'm not sure why I am getting this even though I previously managed to extract the dialogue for the whole mod.