Tool RPGM DazedMTLTool - A tool that provides quality MTL translations using ChatGPT

[dazedanon]

Tested it out and it worked! The skipped line is no more and the translations are properly placed!
What's more is that the names are also fixed and properly done too compared to before when they looked like this:

"Line1": "[Yabuotoko]: “Is it already this late...?”",
  "Line2": "[{ \"Name\": \"Lin\" }]: “Ah! I have to go home now...”",
  "Line3": "[Yabuotoko]: “Well then, if you feel like it, come over to see the cats next time, okay?”",
  "Line4": "[{ \"Name\": \"Lin\" }]: “Yeah ♡”",
  "Line5": "[Yabuotoko]: “See you, Lin-chan!!”",
  "Line6": "[{ \"Name\": \"Lin\" }]: “Onii-chan!!”",

In which I had to manually adjust the prompt to this in the rpgmaker module:
"... translation of the NPC name and just the name only"

Now I don't have to anymore hopefully!
I'll test and see if it still works with my old prompt and if not, I'll add the new prompt guidelines to my current one! ^^

...OH, about the "IndexError: list index out of range" error, that still happens but the files process well and I think it still translates them despite the issue. I'll post this error here incase you figure something out.

Spoiler: Index out of range error

You don't have permission to view the spoiler content. Log in or register now.

This might have to do with the new python 13 version I guess...

EDIT: Okay I tested it!
The skipped "..." issue is fixed no matter which prompt I use!
The new prompt.example you shared actually fixed the name issue so I'll intergrate the changes into my current prompt!
Thanks for helping thus far DazedAnon! ^_^

Yeah maybe a Python 3.13 issue. I'm on 3.12.

 

[Cutefunniness]

Want to test python 3.13? I believe you can use multiple versions of python with the "py" command.
If not than nevermind heh, I probably could install python 12 just in case.

 

[dazedanon]

Want to test python 3.13? I believe you can use multiple versions of python with the "py" command.
If not than nevermind heh, I probably could install python 12 just in case.

Same files right?

 

[Cutefunniness]

Same files right?

Same same, yup!
Test them out if ya want heh.

 

[Jair4x]

Did GitHub ban you? What happened to the tool and translations you had posted?

 

[Cutefunniness]

Did GitHub ban you? What happened to the tool and translations you had posted?

Ah you can find his git repo here, I don't know what happened to his github but he
updates his project here at gitgud:

You must be registered to see the links

 

[HikariSomnium]

Hello everyone,

I've been following the DazedMTLTool project almost since its inception, and the list of translated games is truly impressive.

dazedanon, and everyone who contributed to the translations, you are truly amazing—thank you!

I’ve long wanted to try translating a small visual novel with the help of ChatGPT, but it’s relatively expensive, especially considering that I might make mistakes in the process, leading to wasted tokens. That’s why I was so excited when I saw information in this thread about using the GPT4Free API. Huge thanks to Cutefunniness for that!

I’ve already done a small test translation using the CSV module, and it worked out! I’ll keep testing and will definitely share the results.

By the way, could someone please share the details of their workflow?
I’m planning to translate a novel on Kirikiri. Maybe I’m not fully familiar with it yet, but as far as I understand, the tool doesn’t allow partial translations. At the same time, there’s a high chance that I’ll need to translate specific parts of a file. How do you handle that?

 

[dazedanon]

Hello everyone,

I've been following the DazedMTLTool project almost since its inception, and the list of translated games is truly impressive.

dazedanon, and everyone who contributed to the translations, you are truly amazing—thank you!

I’ve long wanted to try translating a small visual novel with the help of ChatGPT, but it’s relatively expensive, especially considering that I might make mistakes in the process, leading to wasted tokens. That’s why I was so excited when I saw information in this thread about using the GPT4Free API. Huge thanks to Cutefunniness for that!

I’ve already done a small test translation using the CSV module, and it worked out! I’ll keep testing and will definitely share the results.

By the way, could someone please share the details of their workflow?
I’m planning to translate a novel on Kirikiri. Maybe I’m not fully familiar with it yet, but as far as I understand, the tool doesn’t allow partial translations. At the same time, there’s a high chance that I’ll need to translate specific parts of a file. How do you handle that?

For kirikiri games, its mostly about looking at how the .ks files are formatted and creating a regex that will properly extract the text you need to translate.

The tool will handle the rest. (Usually). Check the kirikiri module and look at the regex that is already there in the code and replace it with one that works for your game.

 

[HikariSomnium]

Thank you! I'll keep trying and figuring it out.

 

[dazedanon]

Thank you! I'll keep trying and figuring it out.

You can swing by the Discord and ask questions in the tools channel too. Im a bit busy so might not respond immediately but others can likely help too.

 

[taodentumienque]

Here you go! I'll give you not just the system.json but one of the Map###.json files that has the "「………。」" text too.
Be sure to check the json to see if that text is still there and another "..." appears right after it!





Ah I remember you, you're the one who wrote in a previous post if this supported 18+!
Curious to try DazedMTL eh? Well to answer your question if there's a free to use this...

Yes yes! There's a free way to use this, I've written a couple posts about so check this post first!
https://f95zone.to/threads/dazedmtl...anslations-using-chatgpt.155635/post-15225135

Make sure to follow my instructions and everything should work A-okay and if not, just tell me and I'll try my best to help ya out!

Hi
I follow your post
After run g4f api, i get this error
Do you know how to fix ?

 

[Cutefunniness]

Hi
I follow your post
After run g4f api, i get this error
Do you know how to fix ?

Ah I see, seems there's a missing package for flask!
I wonder why...ah I see what I messed up, I'll help you out!
Try this command and it should install all the requirements needed for g4f api and more!
pip install -U g4f[all]

This will upgrade and provide the requirements needed!
I didn't realize that the command "pip install -U g4f[api]" is just for installing only the requirements needed for the api but also installs all of g4f too!
I'm going to have to update my comment with this realization alright...my bad.

Anyways this should fix it, this will also allow you to use all of g4f too so you can use the gui (web ui where you can use gpt and other providers directly)
and such too, read the github link for more info if you're curious.

Anyways hope this helps! ^^

 

[taodentumienque]

Ah I see, seems there's a missing package for flask!
I wonder why...ah I see what I messed up, I'll help you out!
Try this command and it should install all the requirements needed for g4f api and more!
pip install -U g4f[all]

This will upgrade and provide the requirements needed!
I didn't realize that the command "pip install -U g4f[api]" is just for installing only the requirements needed for the api but also installs all of g4f too!
I'm going to have to update my comment with this realization alright...my bad.

Anyways this should fix it, this will also allow you to use all of g4f too so you can use the gui (web ui where you can use gpt and other providers directly)
and such too, read the github link for more info if you're curious.

Anyways hope this helps! ^^

seem ok now

 

[taodentumienque]

Ah I see, seems there's a missing package for flask!
I wonder why...ah I see what I messed up, I'll help you out!
Try this command and it should install all the requirements needed for g4f api and more!
pip install -U g4f[all]

This will upgrade and provide the requirements needed!
I didn't realize that the command "pip install -U g4f[api]" is just for installing only the requirements needed for the api but also installs all of g4f too!
I'm going to have to update my comment with this realization alright...my bad.

Anyways this should fix it, this will also allow you to use all of g4f too so you can use the gui (web ui where you can use gpt and other providers directly)
and such too, read the github link for more info if you're curious.

Anyways hope this helps! ^^

Thank you
I success translate a game by g4f
I want to ask that
What's the limit of g4f free plan?
How many token can I use it for free ?

 

[HikariSomnium]

I success translate a game by g4f

Did you translate an 18+ game? How big was the text or token volume?

 

[taodentumienque]

Did you translate an 18+ game? How big was the text or token volume?

Yes
I dont know how big the text but g4f seem to be slower than g4p
But it's free so I can accept it

 

[HikariSomnium]

Yes
I dont know how big the text but g4f seem to be slower than g4p
But it's free so I can accept it

Got it, thank you. Have there been any issues with translation, especially regarding content that falls into the 18+ category?

I’m asking because I’ve encountered some oddities with g4f. Despite setting up the list of providers according to the recommendations mentioned earlier, a small portion of the translation came back in Japanese, and some parts seem relatively weak.

For example, "せーし" (seishi) was translated by g4f as "teacher."

I ran another test on an even smaller file, and here are two lines with their translations for comparison:

Source - 中学生になっても身長伸びないっすねぇ♥:

1. Result (Standard "paid" API from OpenAI) - Even in middle school, you haven't grown taller, huh? ♥
2. Result (g4f API) - Even after becoming a middle school student, I’m not growing taller, huh? ♥

It feels like second was translated by 4o mini. In all cases, the .env file specifies the 4o model. For the standard API, it’s "gpt-4o-2024-08-06," while for the g4f API, it’s "gpt-4o" (if you try to select "gpt-4o-2024-08-06" in g4f, it returns an error: 'ModelNotFoundError: Model not found: gpt-4o-2024-08-06'}, 'model': 'gpt-4o-2024-08-06').

Of course, I’m not complaining, given that g4f is free and has enormous potential. I just want to figure out how to achieve better results.

It’s also a bit unclear how to best edit a relatively large file—for instance, one with [Input: 545039][Output: 1359465] tokens—when some parts of the translation are missing or entirely incorrect.

As I understand it, reattempting the translation of this file will result in similar outcomes, but with errors in different sections.

If anyone shares their experience, I’d be grateful (I’ll probably ask around on Discord later).

 

[Cutefunniness]

Thank you
I success translate a game by g4f
I want to ask that
What's the limit of g4f free plan?
How many token can I use it for free ?

The only limit is if the providers are available for use, otherwise…there’s no limit. G4F has no such thing as a free plan, it uses providers (sites that use gpt and such) to send the content to and back.

It’s indeed free, and it’s not perfect but works well! Yup!

 

[Cutefunniness]

Got it, thank you. Have there been any issues with translation, especially regarding content that falls into the 18+ category?

I’m asking because I’ve encountered some oddities with g4f. Despite setting up the list of providers according to the recommendations mentioned earlier, a small portion of the translation came back in Japanese, and some parts seem relatively weak.

For example, "せーし" (seishi) was translated by g4f as "teacher."

I ran another test on an even smaller file, and here are two lines with their translations for comparison:

Source - 中学生になっても身長伸びないっすねぇ♥:

1. Result (Standard "paid" API from OpenAI) - Even in middle school, you haven't grown taller, huh? ♥
2. Result (g4f API) - Even after becoming a middle school student, I’m not growing taller, huh? ♥

It feels like second was translated by 4o mini. In all cases, the .env file specifies the 4o model. For the standard API, it’s "gpt-4o-2024-08-06," while for the g4f API, it’s "gpt-4o" (if you try to select "gpt-4o-2024-08-06" in g4f, it returns an error: 'ModelNotFoundError: Model not found: gpt-4o-2024-08-06'}, 'model': 'gpt-4o-2024-08-06').

Of course, I’m not complaining, given that g4f is free and has enormous potential. I just want to figure out how to achieve better results.

It’s also a bit unclear how to best edit a relatively large file—for instance, one with [Input: 545039][Output: 1359465] tokens—when some parts of the translation are missing or entirely incorrect.

As I understand it, reattempting the translation of this file will result in similar outcomes, but with errors in different sections.

If anyone shares their experience, I’d be grateful (I’ll probably ask around on Discord later).

Ah have you seen my posts about editing the providers list a while back! In case you don’t know, the newest updates of g4f fixed and made the providers I disabled work and not cause any issues.

Make sure to update g4f whenever you can and tro out blackbox/chatgptes as they work better compared to before.

As for some lines reporting back in Japanese or 18+ lines in general, the 18+ lines may tend to require trial and error to get the providers to respond with the translated text while the lines reporting back in Japanese…that happened a lot frequently with an older version of g4f where blackbox wasn’t responding to prompts properly.

But there’s also a chance that a provider might respond in Japanese randomly…but for me I haven’t encountered that as much as I did in the older g4f updates.

Now about the translation of providers being like 4o-mini, there’s a chance that the translation will vary from time to time depending on the provider. I suspect chatgptEs to be like 4o-mini, as for others I’m not sure but you can try disabling that if you want to get a better response.

As for certain words translating differently, be sure to modify the vocab.txt or manually edit the translations for anything that seems off. You can also retranslate a line that seems off by manually removing said translation from the translated file.

Modify your prompt if needed too.

Oh I like to ask if you don’t mind, What game are trying to translate? Is it RPGMaker or something else? Just curious! I ‘ve been translating games of various engine like realive using translators++ csv format along with RPGMaker MVMZ game, one of which I had an issue which dazed helped me with resolving. Strange encoding was in the .json files, that and g4f would respond with improper formatted names when translating in which thr fix was an updated prompt that dazed kindly shared and I added to my already custom prompt.txt message. You can find the posts in the last page or so!

Yup, if there’s any other questions I’ll try to help! Otherwise…wish you the best on translating games. ^^

 

[HikariSomnium]

Ah have you seen my posts about editing the providers list a while back! In case you don’t know, the newest updates of g4f fixed and made the providers I disabled work and not cause any issues.

Make sure to update g4f whenever you can and tro out blackbox/chatgptes as they work better compared to before.

As for some lines reporting back in Japanese or 18+ lines in general, the 18+ lines may tend to require trial and error to get the providers to respond with the translated text while the lines reporting back in Japanese…that happened a lot frequently with an older version of g4f where blackbox wasn’t responding to prompts properly.

But there’s also a chance that a provider might respond in Japanese randomly…but for me I haven’t encountered that as much as I did in the older g4f updates.

Now about the translation of providers being like 4o-mini, there’s a chance that the translation will vary from time to time depending on the provider. I suspect chatgptEs to be like 4o-mini, as for others I’m not sure but you can try disabling that if you want to get a better response.

As for certain words translating differently, be sure to modify the vocab.txt or manually edit the translations for anything that seems off. You can also retranslate a line that seems off by manually removing said translation from the translated file.

Modify your prompt if needed too.

If I’m not mistaken, the version of g4f I’m using is the latest. But just in case, I’ll try updating it. The information you shared here earlier was very helpful. I configured the providers based on your advice. It’s great when there are people willing to help and share useful information (°◡° )

I’ll be testing different providers. For now, I’m thinking of creating a test file for translation and using it to find the most reliable provider. If I manage to find one, I’ll use it to translate in relatively small portions. Although I do have the option to use a paid API, the amount deposited in the account isn’t very large, so I’ll try to save it.

Then again, manual edits are quite difficult and time-consuming, so I might go for the paid option if it turns out to be significantly easier.

Oh I like to ask if you don’t mind, What game are trying to translate? Is it RPGMaker or something else? Just curious! I ‘ve been translating games of various engine like realive using translators++ csv format along with RPGMaker MVMZ game, one of which I had an issue which dazed helped me with resolving. Strange encoding was in the .json files, that and g4f would respond with improper formatted names when translating in which thr fix was an updated prompt that dazed kindly shared and I added to my already custom prompt.txt message. You can find the posts in the last page or so!

It’s a bit embarrassing (⁄ ⁄•⁄ω⁄•⁄ ⁄) but I want to translate "Senkou no Rumina Zenpen Sennyuu! Sei Alucard Jogakuen" and probably "Senkou no Rumina Kouhen Midara na Mesu Juu no Maguwari". Essentially, it’s one game, just split into two parts.

The game was created using the KiriKiri engine. I’ve already exported a CSV file via Translator++ for subsequent translation using DazedMTLTool and, overall, I’ve checked that there are no unsolvable issues with “injecting” the translated file. Now I need to decide whether to use g4f or a "paid API" and figure out the best way to split the file into chunks for batch translation (as far as I understand, Translator++ doesn’t have this feature). It seems I’ll need to split the CSV file manually or with a script, or perhaps it’s better to divide the original file before sending it to Translator++ for subsequent export.

I’ll definitely share the details once it’s all done.

Yup, if there’s any other questions I’ll try to help! Otherwise…wish you the best on translating games. ^^

Thank you so much!♥