apply button? If you're talking about the API Key it is automatically applied once you click any of the generate buttons. You need an elevenlabs account to access the Elevenlabs API. Your api key can be found in your Elevenlabs Profile tab.
1) Open your mod in Xedit 2) right click on your mods name in the side window and click apply script. 3) select the export dialogue script 4) let the script run and save the text file to an easy to remember place 5) import to google sheets or excel and convert to csv (Default format is a tab delimited txt file) 6) Place filename in first column 7) place response text in second column (You can delete any other columns they don't matter) 8) Save as csv file
Did you check if it's possible to include foreign characters yet? As far as I'm aware a good econding should be UTF-8 and my guess is that the tool converts the text to something like Windows-1250.
I got some characters expiring soon and really wanna voice my mod with this tool.
I haven't had any time to work on the tool lately. As it stands now special/foreign characters can not be used in file names. The tool will recognize them for tts input to elevenlabs, but the csv batch generate function will not. It may be possible to work around this problem by ensuring the csv file is saved in a utf-8 encoded format already, but I haven't tested that, so I'm not sure.
All righty. I tried different encodings for my CSV file, even tried to make it read-only but it doesn't seem to have any effect. For now us Germans can still use the tool to some extent by automatically removing any line containing special characters.
OK. It's gonna take a bit of creativity to incorporate it into my typical workflow (I usually generate lines first before putting them in the CK, since I have scripts to handle .lip + .fuz and creating the plugin records) but it seems to work well enough! And it saves me the effort of writing my own python. :D
Awesome tool, endorsed! Works for pretty much every game out there.
One question, which might has nothing to do with your tool. When generating speech in the tool itself, it works for special characters we have in Germany (ä, ö, ü). When using batch generate, it replaces those special characters with some gibberish, even though the csv file still shows them correctly.
Is there any encoding being done by the tool, or did I fail to prepare the csv correctly?
The problem would be that I didn't take into account special characters when writing the code. The tool is a python based application, so special encoding is required for special characters. I'll see if I can add a fix for this with the next update of the app.
It may work with the direct speech generation simply because it's sending the raw data to elevenlabs, but the csv batch generation has a bit more processing within the script itself. It should be a relatively simple fix so I'll see what I can do.
I finally decided to use your program seriously and noticed that it is built in Python. I suppose you packaged it using PyInstaller or another tool. Generally, these programs generate false positives when creating executables. You can use others like cx_Freeze, which tends to package programs without these errors, or utilize other programs that repair the errors.Upon a simple review, I noticed that the code language is not defined.
Generally, it should be set from 0 (neutral) to 1033 (English) to avoid false positive issues. Additionally, I recommend using VirusTotal, which is a website that allows you to upload your program and have it scanned by more than 70 antivirus programs in seconds. The antivirus programs use this as a data bank, which adds more credibility to the program.
Spoiler:
Show
Sorry for the long explanation, but the same thing used to happen to me when I created programs with those horrible false positives. They can be solved, but even with the change of language code, they will still appear—it's just a Python thing.
Back to the program, it looks great, but my cat brain still doesn't fully understand how it works. But those are just details. I still don't fully understand how to use it, but it looks promising. A simple question: how is the Voice-to-Voice feature used? Also, excuse my ignorance regarding ElevenLabs; the voice models of the game’s NPCs are not there. Do I need to upload something or generate it myself? For now, thank you for the program. I was already thinking about creating one myself, but this will help me a lot if I understand how to use it correctly.
Thanks for the advice, it's my first python application so pyinstaller was pretty much the first method I found to freeze python code with. I appreciate the advice and I'll definitely keep it in mind.
I actually didn't + still don't know what you mean by "code language undefined" 😅
All I know is it does what I wanted it to do somewhere in my spaghetti code. 🤓
Glad you were able to figure out how to use it and ty for the pointers.
So I have the tool now, and I think I get how it works for TTS... but for STS, is the only way to generate STS output through using a microphone? I notice that there is a STS input folder, is there some way I could put some audio files into the STS Input folder, and then have the tool generate STS output for me?
I hadn't thought of using pre-recorded audio for sts. There is no current way to do that with my tool. you can still do it via elevenlabs website. I'll see if i can add that to the tool when I get some more free time.
There are some really interesting use cases that would make STS batch generation worthwhile, like adding more generic voices. Rather than TTS, STS is more emotive if the input voice is good, and is better with accents, intonation generally.
I’d stoked if you managed to add that into the tool!
LOVE LOVE LOVE the new version of your tool, love it so much! Thank you, you have no idea how many hours of renaming all these dreadful files you have saved me many headaches and I've gotten so many voices and lines generated. It has massively cut my work load by a huge margin, got probably 8-16 hours worth of renaming voice lines done, yay! Thank you so much for such an easy to use and nifty program. New version works great, glad it got approved! I was worried there for a while with it being flagged. Was about to write you and suggest maybe uploading to Github and link from Nexus or something. Some modders have had to do that since Nexus gets false positives all the time. Make a mod page, then link to an offisite website, that's what many do. Anyway, glad to see the new version and thanks again!
I'm a moderator for the Discord of the mod Conquest of Skyrim, it's a large ambitious mod fully voiced by Eleven Labs AI, where the player creates their own faction and government that rivals the Empire and Stormcloaks, where you appoint Jarls, Stewards and Housecarls for all the holds and become elected High King or High Queen. And with your tool I can update my tutorial for adding custom voice support to the mod, then anybody can add custom voices for their follower mods into it easily. If you haven't heard of the mod it's pretty cool. Conquest of Skyrim at Skyrim Special Edition Nexus - Mods and Community (nexusmods.com)
In less than 2 months the mod is getting a massive new overhaul and update, we will be able to have royal feasts, and a settlement building system like Fallout 4, diplomacy and spies/assassins, and things will be alot more innovative and immersive on how we deal with the world and our enemy jarls. There will also be a rebellion system where our own people or appointed jarls can turn against us if we don't take care of them.. Angry mobs, and full on uprisings. Anyway feel free to join our server if you ever want to. We have a really good community of fellow Skyrim modders.
I'm happy to hear that my tool has helped you with your mod. Conquest of Skyrim is looking very epic! Hopefully now with my tool you'll be able to spend more time creating an awesome mod, and less time renaming voice files!
I'm actually not the mod creator, that would be Galvatron but I do help out with it with additional voice files, and made custom banners and tent meshes for the mod. Indeed, less time renaming files is nice, because it could be tedious when you have to do so many.
I just wish there was a way to do the renaming part your tool does for all my old generated files I haven't renamed yet, since those files are all already generated from before your tool, I unfortunately have to rename those manually. I created a Python script a while back but it only seems to rename more than half of the lines and not all. Your tool seems to do every single one. My python script ignores the files from the CSV that end with a question mark, so those sentences/lines get skipped and can't figure out how to get it working. I want to complete this python script so people that record with their own voice would have an easier method instead of renaming many files.
I did some fiddling with my generator and made a few changes. Including Better layout, Speech to speech integration, removal of the need for windows media player, and hopefully helpful to you: adding a converter to convert text to and from DBVO's dialogue format via a csv file. (in the same format the generator uses Filename,"Response Text") It also addresses the problem you were having by converting "?" to "_" (which is what DBVO does)
Really? Wow! You're a genius! And thank you so much for doing that for me, that sounds very beneficial and will help out not only myself but many modders.
I look forward to trying it out later. 🙂
________________ Oh I did also have a couple things I would recommend if you ever consider them is a Lip File Generator, and an Auto Fuz maker after all Lip and wavs are together to fuz them all, generating fuz files from all Lip and wavs
the creation kit already has a built in batch lip file generation function (and so does dbvo's generator and runalip) and I use Yakitori for Conversion to fuz. I don't really want to dive into the complexities of encoding files into Bethesda's proprietary formats when there are already resources that do a good job for that purpose.
40 comments
If you're talking about the API Key it is automatically applied once you click any of the generate buttons.
You need an elevenlabs account to access the Elevenlabs API. Your api key can be found in your Elevenlabs Profile tab.
2) right click on your mods name in the side window and click apply script.
3) select the export dialogue script
4) let the script run and save the text file to an easy to remember place
5) import to google sheets or excel and convert to csv (Default format is a tab delimited txt file)
6) Place filename in first column
7) place response text in second column
(You can delete any other columns they don't matter)
8) Save as csv file
As far as I'm aware a good econding should be UTF-8 and my guess is that the tool converts the text to something like Windows-1250.
I got some characters expiring soon and really wanna voice my mod with this tool.
I tried different encodings for my CSV file, even tried to make it read-only but it doesn't seem to have any effect.
For now us Germans can still use the tool to some extent by automatically removing any line containing special characters.
Thanks! And endorsed!
Works for pretty much every game out there.
One question, which might has nothing to do with your tool.
When generating speech in the tool itself, it works for special characters we have in Germany (ä, ö, ü).
When using batch generate, it replaces those special characters with some gibberish, even though the csv file still shows them correctly.
Is there any encoding being done by the tool, or did I fail to prepare the csv correctly?
It may work with the direct speech generation simply because it's sending the raw data to elevenlabs, but the csv batch generation has a bit more processing within the script itself. It should be a relatively simple fix so I'll see what I can do.
Generally, it should be set from 0 (neutral) to 1033 (English) to avoid false positive issues. Additionally, I recommend using VirusTotal, which is a website that allows you to upload your program and have it scanned by more than 70 antivirus programs in seconds. The antivirus programs use this as a data bank, which adds more credibility to the program.
Back to the program, it looks great, but my cat brain still doesn't fully understand how it works.
But those are just details.
I still don't fully understand how to use it, but it looks promising. A simple question: how is the Voice-to-Voice feature used? Also, excuse my ignorance regarding ElevenLabs; the voice models of the game’s NPCs are not there. Do I need to upload something or generate it myself?For now, thank you for the program. I was already thinking about creating one myself, but this will help me a lot if I understand how to use it correctly.Edit: I already understood it, thank you,
I actually didn't + still don't know what you mean by "code language undefined" 😅
All I know is it does what I wanted it to do somewhere in my spaghetti code. 🤓
Glad you were able to figure out how to use it and ty for the pointers.
This looks incredible, can’t wait to try it out!
I do want to ask, you mention on the page that there is speech to speech integration, how would that work?
Thanks!
It uses your default microphone.
Press start listening or use the hotkey.
If you have a filename entered it will save to that filename, otherwise it will name your output with a timestamp.
Input audio is saved to an sts inputs folder in the destination directory so you can regenerate audio from the log.
I’ve jumped the gun a bit since I’m at work atm, will try this out when I get home.
If I can get this to work I’ve got some very exciting things planned lol
There are some really interesting use cases that would make STS batch generation worthwhile, like adding more generic voices. Rather than TTS, STS is more emotive if the input voice is good, and is better with accents, intonation generally.
I’d stoked if you managed to add that into the tool!
Anyway, glad to see the new version and thanks again!
I'm a moderator for the Discord of the mod Conquest of Skyrim, it's a large ambitious mod fully voiced by Eleven Labs AI, where the player creates their own faction and government that rivals the Empire and Stormcloaks, where you appoint Jarls, Stewards and Housecarls for all the holds and become elected High King or High Queen. And with your tool I can update my tutorial for adding custom voice support to the mod, then anybody can add custom voices for their follower mods into it easily. If you haven't heard of the mod it's pretty cool. Conquest of Skyrim at Skyrim Special Edition Nexus - Mods and Community (nexusmods.com)
In less than 2 months the mod is getting a massive new overhaul and update, we will be able to have royal feasts, and a settlement building system like Fallout 4, diplomacy and spies/assassins, and things will be alot more innovative and immersive on how we deal with the world and our enemy jarls. There will also be a rebellion system where our own people or appointed jarls can turn against us if we don't take care of them.. Angry mobs, and full on uprisings. Anyway feel free to join our server if you ever want to. We have a really good community of fellow Skyrim modders.
I just wish there was a way to do the renaming part your tool does for all my old generated files I haven't renamed yet, since those files are all already generated from before your tool, I unfortunately have to rename those manually. I created a Python script a while back but it only seems to rename more than half of the lines and not all. Your tool seems to do every single one. My python script ignores the files from the CSV that end with a question mark, so those sentences/lines get skipped and can't figure out how to get it working. I want to complete this python script so people that record with their own voice would have an easier method instead of renaming many files.
Here's my mod page though, in my files I have my py files.
Conquest Of Skyrim - Additional Voices Patch at Skyrim Special Edition Nexus - Mods and Community (nexusmods.com)
I look forward to trying it out later. 🙂
________________
Oh I did also have a couple things I would recommend if you ever consider them is a Lip File Generator, and an Auto Fuz maker after all Lip and wavs are together to fuz them all, generating fuz files from all Lip and wavs
but I'll consider it for the future... who knows.