If you guys encounter any voices that you feel are below the standard set by the other voices, let me know! I'm not against releasing some updates redoing weaker lines.
Edit; Make sure you install BOTH update files or two characters will have a few incorrect lines.
This mod actively has actively made me consider specifically disallowing AI revoices for all of my mods, and I'd recommend other mod authors do the same. Just a gross thing to do.
You do you. I mean that’s your choice as author if you want to disallow that. The author of WWM had open permissions so I don’t see anything wrong with it. I just wanted voices that meshed better with the vanilla game. You may disagree with the ethics of AI, thats your choice, but calling it "Gross" is a bit dramatic...
This over the top aversion of AI seems absurd to me. I slightly understand the fear of having a profession that could be 'replaced' with AI but it could also give independent creators without much budget the opportunity to see their visions come to life. I think that's especially applicable in the modding scene, where people create for free. Corporations specialising in visual/audio media should not use AI to replace people, but most amateurs don't have the money and/or connections to access quality VAs and artists. And if you're creating something for nothing, I don't feel like you owe anyone else work.
Getting VAs is possibly the easiest part of mod making.
And I do find taking someone else's mod and overwriting the hard work of others with a robot to be gross. Why mod at all if you're just gonna take the heart and passion out? Just removing the earnest efforts of the community and replacing it with slop.
You clearly don’t understand the We Want More mod. The original author didn’t hire voice actors, something you claim is “the easiest part of mod making.” Instead, he voiced 30+ new characters himself, along with new lines for vanilla NPCs. That’s a massive amount of work, and I respect the dedication that went into it. However, the new lines for vanilla characters didn’t mesh well. Hearing their voices suddenly change mid-conversation didn’t make sense and broke the immersion for me. It was clear that some voices got more love and attention than others, and on top of that, the mic quality was pretty poor across the board. The voice acting was literally the one flaw holding We Want More back.
You’re acting like I took pro-level voice acting like Willow's or Steven Randall's and revoiced it with AI. That’s not the case here. I spent over a dozen hours fine-tuning every line in the mod to create voices that I felt meshed better with the vanilla game. I guarantee I spent more time fine-tuning each particular line than the original author did recording each one. I didn’t overwrite his hard work with slop, I overwrote his hard work with my own hard work, crafting voices that match the quality of vanilla.
My goal wasn’t to undermine the original author’s work but to elevate it to a point where it's more enjoyable to my fellow immersion junkies in 2025. Judging by other comments here, from people who actually use We Want More (unlike you) they agree with me. At the end of the day, modding is about improving the game experience in ways that matter to yourself and others. I'm proud of this mod, you can kick rocks.
"I guarantee I spent more time fine-tuning each particular line than the original author did recording each one."
Lol. Lmao, even. How egotistical can you be after telling a robot to do the actual work for you? For someone that claims to hold the original in such high regards, you don't seem to respect the work that went into it much.
Fine-tuning AI-generated voices isn’t just pressing a button and calling it a day. It’s hours of adjusting tone, pitch, timing, delivery, and often splicing multiple lines together with audio editing software to ensure the dialogue feels natural and blends seamlessly with the vanilla game. If you think that’s effortless, I commend you for wearing your ignorance like a badge of honor.
I respect the work that went into the original mod, which is why I put in the time to address what I saw as its only real flaw. If I didn’t respect his work, why would I have bothered doing this? I made this mod because I loved what he did, but like anyone, I can enjoy some aspects of something while disliking others. I was about 6 years old when I learned this.
My mod isn’t about tearing down the original. It’s about elevating it for those of us who care about immersion. If that offends you, which it clearly does, please seek therapy.
Might as well hate on sound remixers, audio tuners, recording programs, or even audio speakers, because they offload your share of the work onto a machine. If a mod doesn't have the voice actors directly coming to my house it's slop and I refuse to use it!
Having the mod VA's come over to your house for a nice, properly narrated session of new vegas sounds like a regular rich furry activity. I feel like it has happened at some point, and we just didn't get to see it happen cause it wasn't documented.
"Getting VAs is possibly the easiest part of mod making." This is a ridiculous thing to say and so disrespectful to talented voice actors. There are astronomically few mods with high quality voice acting. It is, by and away, the hardest thing for most amateur mods to get right.
My guy Danneyo says he'd disallow AI voices in addons to his mods in the future, while none of his mods seem to have any voice lines at ALL. These Neo-Luddites are a tiresome sort.
@TheUncleG Your work here is much appreciated. These old FO3 mods especially need this work. It is true that the og VA(usually on a pre 2013 laptop mic by a modder with no other choice), holds the mods back today, and can make them pretty unplayable. From what I've played of the mod so far, and compared to the original, I'd say this has elevated a very good mod to a very very good mod and made it like part of the game. I imagine the og author would be glad that someone thought enough of his work to put the effort in. I would. And it is effort to do properly what you did here.
Wow, thank you! Your comment sincerely made my day, especially after some of the negative feedback I received on this one. I didn’t realize just how contentious AI voice acting could be! I really appreciate you recognizing the effort that went into this and understanding that my goal wasn’t to diminish the original. Hearing that it makes the mod feel like a natural part of the game is the best compliment I could ask for.
Self righteousness is a very negative trait. Not even justified. Funnier because you make no real quest mod content for anyone to care to AI voice. Only reason anybody would would be to screw with you now that you picked a fight in the comments of an AI voice mod. Acted genuinely disrespectful and made bad faith claims like somehow getting voice actors being so easy. I've talked to people in smaller projects. It is not very easy. Then again, you're a Frontier dev(already a good sign), a large project, where it's more likely for people to flock to try voicing for the mod.
Either way, it's not your call who gets to create AI voice mods. Perhaps read the Nexus policy sometime. :)
Man, I'd be pissed if I sunk a load of effort and enthusiasm into a New Vegas passion project, only to have some turd leave insults in my comment section about how my mod is bad and I should feel bad about making it...
...oh wait, that DID happen. And I ditched my account, started with a fresh one and never got back into Fallout modding. So thank you people like OUtlander3k for being so supportive and encouraging. Imagine how many great mods might exist if you didn't exist? Cheers.
Hey bro this was amazing, i have clipped some of this mod as its my first time playing it and i hope i bring some downloads your way, keep up the great work..
Can you replace the voiceline of the mail carriers in this mod? Great work btw. This ai voice acting just makes it feels more vanilla and lore friendlier. Don't listen to what other people crying. No one has to pay anything to download the mod. Id rather playing with unvoice npc than having a low quality va character
Oh my goodness, this is so good! I should've encountered this mod sooner! Now I want have to replay this mod again just to enjoy the immersion. Thank you so much for this! You have no idea how incredible and well put together this mod is. Not a lot of people are willing to make a mod like this due to dislike for AI content, but I for one, accepts and appreciate this new use of technology on older games.
Oh, and one question, is there a tutorial or instructions out there on how to make this kind of mod, like the modifying and implementing of the AI voices to characters in the intended mod? I already know how to make AI voices since I've played around with open source AI TTS models before.
By the way, if anyone's interested, cloning AI voices locally are now easier and free as long as you have a good NVIDIA GPU, and if you're willing to be a bit technical with AI. No need to pay ElevenLabs anymore, lol (No offense). I mean, take a look at this comparison video.
Sorry, I'm just a bit too excited, so I'm excuse me if I said anything wrong. Once again, thank you very much for your work. It's really inspiring!
Thank you! Your comment made my day! As far as a tutorial, not really. I kinda winged it 😂 Best I can do is give you some bullet points on what I did and you can see if you can apply this to other mods. Here are the (rough) steps:
- Get permission from the original author to do the revoice in the first place (The WWM author had open permissions) - Manually download the mod you are trying to revoice - Open the mod up with a file explorer (The mod may be packaged as a BSA archive, if this is the case you can extract the contents of the BSA with a BSA extractor tool from the Skyrim nexus https://www.nexusmods.com/skyrimspecialedition/mods/974) - Find the voice files the mod uses, in the case of WWM it was under sound>voice>wwm.esm>[next folder is named after the specific character I am revoicing] - Once you are in the character's folder, you'll see some .ogg and .lip files. The .ogg is the audio, and the .lip is the lip syncing. - Open up a new blank folder on your desktop and recreate the sound>voice>wwm.esm>[Character Name] file structure. For the [Character Name] folder, you need to use the same exact name the mod uses to identify the character. For example "Blake"'s file in this mod is called "aablakevoicetype" so the file structure on my desktop would be sound>voice>wwm.esm>aablakevoicetype - Copy the .lip files from "aablakevoicetype" from the original mod to your new "aablakevoicetype" folder - Open up ElevenLabs and drag one of the .ogg voice lines from the original mod into it - Generate (and regenerate) the voice in ElevenLabs until you are happy with it - Once you are happy with the voice, download it as an MP3 - Add the new voice to your newly created character folder (For this example the "aablakevoicetype" folder found within sound>voice>wwm.esm on my desktop) - Name the new MP3 voice line you just made the exact same as the original .ogg file it replaced - Repeat the ElevenLabs steps until you have MP3 voice lines for every line of that characters dialogue in that character folder on your desktop - Download Audacity (free) and start a new project, upload ALL the MP3 voice lines from ElevenLabs and bulk export them as .ogg files to your desktop>sound>voice>wwm.esm>aablakevoicetype folder - Repeat for every character you want to revoice - Once that is done, delete (or move elsehwere) ALL the MP3 files from the desktop>sound>voice>wwm.esm>aablakevoicetype folder. This folder should now contain only .ogg and .lip files. The .ogg files should be your AI revoices - Repeat the last 12 steps for every character you want to revoice - Package the mod with 7zip or Winrar and add it to MO2, making sure it is loaded BELOW the mod you are trying to revoice (so the new mod overwrites the original voices)
Sorry if that was too much and a bit confusing and hard to follow. I haven't touched this mod in 2 months, and the bulk of the work was done 4 months ago, and I wrote all those instructions by memory, so theres a chance I am missing a step. But overall thats how I did it. If you ever are in a position where you're trying to follow these instructions and something is going wrong just reply here and I will try to help you out if I can.
Oh wow, this is definitely much needed. I'll be sure to take my time with all the steps. Considering I also have to learn on how to extract the BSA and using Audacity. All in all, it's a very thorough step-by-step process if you ask me. Just need to tweak something if there's anything else I need.
Anyway, I really appreciate this. This will definitely be a long monumental task, but it's easier now thanks to your guidance. Thank you so much
How did you manage to clone the original voice references, though? It's my first time playing in English so I might be mistaken but and ElevenLabs requires you to verify all voice inputs when utilizing their own cloning feature.
It's just kinda immersion breaking when the NPCs keep calling my chick as dude or sir. Other than that, the AI voice acting is very good. It feels rather like part of the game than an actual mod. I don't get why people hate it if you don't make any profit out of this. Keep up with your good work.
Any way you can make a version that ONLY replaces the vanilla NPCs lines, because this is just disrespectful to voice actors. Like I get doing it for consistency and/or mod compatibility, but damn man, why replace legit voice actors unless the quality is dogshit.
Did you hear the Voices in the original mod? The acting itself wasn't bad, but the mic quality was horrible. Anything above "dogshit" may be your bar, but I like my voices to be on par or at least close to vanilla in terms of sound quality. Nothing disrespectful about it
I've been playing the mod without the AI voices. Mic Quality is not unlistenable in most cases, it's a little bit rough but it's welcome enough, I've heard a LOT worse in much bigger mods. RVC, what you seemed to use here, does not fix mic quality anyways. The only way you could genuinely make them sound on par with vanilla is spending hours and hours manually generating and regenerating the lines in ElevenLabs. (You also may have used the Speech to Speech function in ElevenLabs, which is also RVC layered with ElevenLabs isolation and TTS features. This will still leave weird leveling issues that the original lines have, and it definitely does.) You could theoretically also use a program like izotope rx 8 to adjust these lines to sound on par with vanilla microphones. This is a tricky process but once you have a solid chain, you can stick to that.
"The only way you could genuinely make them sound on par with vanilla is spending hours and hours manually generating and regenerating the lines in ElevenLabs."
This is literally exactly what I did. Probably spent about 12 hours in ElevenLabs if I had to guess, using both the Speech to Speech and Text to Speech feature. I tried to mostly stick to Speech to Speech, but some lines I couldn't get to work and had to resort to text to speech. These are the cases where the lip syncing is the most off. I know the mod isn't perfect, but its alot closer to vanilla sounding than the original mod. Even though the original voices weren't "unlistenable" they were still off enough to mess up my immersion when I met a new NPC. I play Fallout to be immersed, so if a mod takes me out of the game it kinda is a deal breaker for me. But I loved the idea of WWM and wanted new quests in the capital so I went to work on the AI voices. Again, it's not perfect but it is less immersion breaking imo.
Also, to respond to your original comment. You can just install with manager then open up the mod and navigate to sound>voice>wwm.esm and delete all folders except the one for Gob, Jericho, Billy Creel, and Lucas Simms. This should get your desired effect of only adding the AI voices to the vanilla NPCs
Wow this is amazing, really appreciate the effort put into this. Have you ever thought of doing something similar for some of the less quality voices for 3DNPC's also by chance? Totally understand not wanting to because the obvious amount of effort it would take.
This was a lot more work than I was anticipating, so I greatly appreciate it!
I'm not against it, but from what 3DNPC's I've encountered so far (Dot's Diner, The couple in the Brass Lantern, Slasher, and Mortar), I actually thought they had good voice acting. At least good enough to the point that they didn't take me out of the game like the WWM voices did. Which particular characters from that mod were you thinking of?
Overall I agree, about 95% of them have really solid voice acting that ranges from good enough to great. But as I was in my playthrough 1 in particular stood out to me as not being up to par as you can hear their microphone humming in the background. It's Scribe Hammond located in the Science and Technology Archives. It's hard to find because it's an unmarked location near the Museum of History.
Anyways that's the only one I've found in my playthrough so far that seemed quite out of place and clearly a person with a microphone. Unfortunately the actual voice acting is good, just you can hear their mic pretty easily humming.
Alright I spent some time, looking through every NPC added by the FO3 3DNPC's mod (not the NV counterpart, at least not yet). And frankly the only one that seemed of any real issue was Scribe Hammond. Every other NPC was Good to great IMO from listening to them all with regards to mic quality.
Thanks for doing that. I’ve been thinking about it, and since the mod author went out of their way to hire a voice actress (or so it seems), I’d feel weird revoicing it with AI. I’d rather try to find a way to clean up the audio while keeping the original recordings intact
I found a really easy (relatively) and clean way to enhance poor sounding audio, using Adobe Enhance Speech v2. It's free and very simple, just drag and drop the file in essnetially and it comes out alot cleaner and more professional sounding. Only issue is it's 1 file at a time which would take a long time, unless you get the PRO version. However there is a 7 day free trial. https://podcast.adobe.com/enhance#
Nice find!! I will check that out. Fixing Hammond's lines is the next thing on my list after I get the changes to this mod sorted out. Appreciate it, you've been very helpful
62 comments
Edit; Make sure you install BOTH update files or two characters will have a few incorrect lines.
And I do find taking someone else's mod and overwriting the hard work of others with a robot to be gross. Why mod at all if you're just gonna take the heart and passion out? Just removing the earnest efforts of the community and replacing it with slop.
You’re acting like I took pro-level voice acting like Willow's or Steven Randall's and revoiced it with AI. That’s not the case here. I spent over a dozen hours fine-tuning every line in the mod to create voices that I felt meshed better with the vanilla game. I guarantee I spent more time fine-tuning each particular line than the original author did recording each one. I didn’t overwrite his hard work with slop, I overwrote his hard work with my own hard work, crafting voices that match the quality of vanilla.
My goal wasn’t to undermine the original author’s work but to elevate it to a point where it's more enjoyable to my fellow immersion junkies in 2025. Judging by other comments here, from people who actually use We Want More (unlike you) they agree with me. At the end of the day, modding is about improving the game experience in ways that matter to yourself and others. I'm proud of this mod, you can kick rocks.
Lol. Lmao, even. How egotistical can you be after telling a robot to do the actual work for you? For someone that claims to hold the original in such high regards, you don't seem to respect the work that went into it much.
I respect the work that went into the original mod, which is why I put in the time to address what I saw as its only real flaw. If I didn’t respect his work, why would I have bothered doing this? I made this mod because I loved what he did, but like anyone, I can enjoy some aspects of something while disliking others. I was about 6 years old when I learned this.
My mod isn’t about tearing down the original. It’s about elevating it for those of us who care about immersion. If that offends you, which it clearly does, please seek therapy.
If a mod doesn't have the voice actors directly coming to my house it's slop and I refuse to use it!
I feel like it has happened at some point, and we just didn't get to see it happen cause it wasn't documented.
Your work here is much appreciated. These old FO3 mods especially need this work. It is true that the og VA(usually on a pre 2013 laptop mic by a modder with no other choice), holds the mods back today, and can make them pretty unplayable. From what I've played of the mod so far, and compared to the original, I'd say this has elevated a very good mod to a very very good mod and made it like part of the game. I imagine the og author would be glad that someone thought enough of his work to put the effort in. I would. And it is effort to do properly what you did here.
I hope you do more. Effort appreciated.
Wow, thank you! Your comment sincerely made my day, especially after some of the negative feedback I received on this one. I didn’t realize just how contentious AI voice acting could be! I really appreciate you recognizing the effort that went into this and understanding that my goal wasn’t to diminish the original. Hearing that it makes the mod feel like a natural part of the game is the best compliment I could ask for.
Either way, it's not your call who gets to create AI voice mods. Perhaps read the Nexus policy sometime. :)
@Outlander3k
...oh wait, that DID happen. And I ditched my account, started with a fresh one and never got back into Fallout modding. So thank you people like OUtlander3k for being so supportive and encouraging. Imagine how many great mods might exist if you didn't exist? Cheers.
Great work btw. This ai voice acting just makes it feels more vanilla and lore friendlier. Don't listen to what other people crying. No one has to pay anything to download the mod. Id rather playing with unvoice npc than having a low quality va character
Oh, and one question, is there a tutorial or instructions out there on how to make this kind of mod, like the modifying and implementing of the AI voices to characters in the intended mod? I already know how to make AI voices since I've played around with open source AI TTS models before.
By the way, if anyone's interested, cloning AI voices locally are now easier and free as long as you have a good NVIDIA GPU, and if you're willing to be a bit technical with AI. No need to pay ElevenLabs anymore, lol (No offense). I mean, take a look at this comparison video.
Sorry, I'm just a bit too excited, so I'm excuse me if I said anything wrong. Once again, thank you very much for your work. It's really inspiring!
- Get permission from the original author to do the revoice in the first place (The WWM author had open permissions)
- Manually download the mod you are trying to revoice
- Open the mod up with a file explorer (The mod may be packaged as a BSA archive, if this is the case you can extract the contents of the BSA with a BSA extractor tool from the Skyrim nexus https://www.nexusmods.com/skyrimspecialedition/mods/974)
- Find the voice files the mod uses, in the case of WWM it was under sound>voice>wwm.esm>[next folder is named after the specific character I am revoicing]
- Once you are in the character's folder, you'll see some .ogg and .lip files. The .ogg is the audio, and the .lip is the lip syncing.
- Open up a new blank folder on your desktop and recreate the sound>voice>wwm.esm>[Character Name] file structure. For the [Character Name] folder, you need to use the same exact name the mod uses to identify the character. For example "Blake"'s file in this mod is called "aablakevoicetype" so the file structure on my desktop would be sound>voice>wwm.esm>aablakevoicetype
- Copy the .lip files from "aablakevoicetype" from the original mod to your new "aablakevoicetype" folder
- Open up ElevenLabs and drag one of the .ogg voice lines from the original mod into it
- Generate (and regenerate) the voice in ElevenLabs until you are happy with it
- Once you are happy with the voice, download it as an MP3
- Add the new voice to your newly created character folder (For this example the "aablakevoicetype" folder found within sound>voice>wwm.esm on my desktop)
- Name the new MP3 voice line you just made the exact same as the original .ogg file it replaced
- Repeat the ElevenLabs steps until you have MP3 voice lines for every line of that characters dialogue in that character folder on your desktop
- Download Audacity (free) and start a new project, upload ALL the MP3 voice lines from ElevenLabs and bulk export them as .ogg files to your desktop>sound>voice>wwm.esm>aablakevoicetype folder
- Repeat for every character you want to revoice
- Once that is done, delete (or move elsehwere) ALL the MP3 files from the desktop>sound>voice>wwm.esm>aablakevoicetype folder. This folder should now contain only .ogg and .lip files. The .ogg files should be your AI revoices
- Repeat the last 12 steps for every character you want to revoice
- Package the mod with 7zip or Winrar and add it to MO2, making sure it is loaded BELOW the mod you are trying to revoice (so the new mod overwrites the original voices)
Sorry if that was too much and a bit confusing and hard to follow. I haven't touched this mod in 2 months, and the bulk of the work was done 4 months ago, and I wrote all those instructions by memory, so theres a chance I am missing a step. But overall thats how I did it. If you ever are in a position where you're trying to follow these instructions and something is going wrong just reply here and I will try to help you out if I can.
Anyway, I really appreciate this. This will definitely be a long monumental task, but it's easier now thanks to your guidance. Thank you so much
RVC, what you seemed to use here, does not fix mic quality anyways. The only way you could genuinely make them sound on par with vanilla is spending hours and hours manually generating and regenerating the lines in ElevenLabs. (You also may have used the Speech to Speech function in ElevenLabs, which is also RVC layered with ElevenLabs isolation and TTS features. This will still leave weird leveling issues that the original lines have, and it definitely does.)
You could theoretically also use a program like izotope rx 8 to adjust these lines to sound on par with vanilla microphones. This is a tricky process but once you have a solid chain, you can stick to that.
This is literally exactly what I did. Probably spent about 12 hours in ElevenLabs if I had to guess, using both the Speech to Speech and Text to Speech feature. I tried to mostly stick to Speech to Speech, but some lines I couldn't get to work and had to resort to text to speech. These are the cases where the lip syncing is the most off. I know the mod isn't perfect, but its alot closer to vanilla sounding than the original mod. Even though the original voices weren't "unlistenable" they were still off enough to mess up my immersion when I met a new NPC. I play Fallout to be immersed, so if a mod takes me out of the game it kinda is a deal breaker for me. But I loved the idea of WWM and wanted new quests in the capital so I went to work on the AI voices. Again, it's not perfect but it is less immersion breaking imo.
Also, to respond to your original comment. You can just install with manager then open up the mod and navigate to sound>voice>wwm.esm and delete all folders except the one for Gob, Jericho, Billy Creel, and Lucas Simms. This should get your desired effect of only adding the AI voices to the vanilla NPCs
I'm not against it, but from what 3DNPC's I've encountered so far (Dot's Diner, The couple in the Brass Lantern, Slasher, and Mortar), I actually thought they had good voice acting. At least good enough to the point that they didn't take me out of the game like the WWM voices did. Which particular characters from that mod were you thinking of?
Anyways that's the only one I've found in my playthrough so far that seemed quite out of place and clearly a person with a microphone. Unfortunately the actual voice acting is good, just you can hear their mic pretty easily humming.
https://www.youtube.com/watch?v=nuYNQmBnlwg