FastChat is a local model with with 90% ChatGPT quality. The most important is free to use.
And FastChat has OpenAI-Compatible RESTful APIs & SDK.
So it's helpful to add a option to replace openai to fastchat api.
Thanks!

Describe the problem you confuse

That is to say, it is not just used to translate srt
Can also support ordinary TXT translation?

Majority of the outcome is not translated, just copied original text (input Italian sub –> target English.

I'm receiving the prompt "The model: gpt-4-0314" after clicking the Translate button.

Can u show me a solution to get a free Open AI API key for the Subtitle Translator ?

Deleted

Which method you use?

• gpt-4-0314
• gpt-3.5-turbo
• gpt-3.5-turbo-economy

Provide your subtitle file

Dungeons.And.Dragons.Honor.Among.Thieves.English-WWW.MY-SUBS.CO.srt.zip

Target language

Thai

Additional notes

None

Describe the bug

Gets a few percent in and throws TypeError: Cannot read properties of undefined (reading 'data')

To Reproduce

Normal use as per instructions

我尝试不同的音频转字幕得到的文件去尝试

Rate limit后，再次翻译会从头重新开始，已经翻译过的部分又会再次翻译。

Describe the problem you confuse

I bought the OpenAI api service from a reseller, but I need to work with the reseller's host to call the api properly. I hope the author can add an OpenAI API Host settings.I don't know anything about code🥲

The translation speed is a bit slow. Can it be optimized?

It seems that the token consumption is a bit high now. Can it be reduced?

Thank you.

I entered like this it is going on but this is the proper way? I want to translate into multiple languages at 1 run. Or any queue system?

Turkish, Spanish

i am hoping it is :/

将英日混合的文本翻译为英文时，一些日文不会被翻译，会维持原文不变，点击刷新后才会翻译成英文。

Which method you use?

• gpt-3.5-turbo

Provide your subtitle file

https://drive.google.com/file/d/1uyr2in1668awf688ye_46Ul17ZpTb21Q/view?usp=share_link

Target language

Brazilian Portuguese

Additional notes

Tradução de um donghua.

Describe the bug

I will bring the prints to show the problem:

Describe the problem you confuse

I got chatGPT plus

it can translate entire subtitle but you have to split it into like 1 minute frames and type again and again

which takes huge time

can you modify it so that we can use chat GPT plus subscription not API?

We need that.

hello, thanks for this wonderful app.

Bug: I found that some sentences were not translated when I used it, and the API returned the original text. Can you add the function of single sentence retry?
Sometimes the translation result returned by the API contains the translation of the next sentence. I understand that context helps with translation. Some sentences filter out such results while others do not. I hope it can be fixed.

Thanks again.

Here are some quality of life adds I and I think others would like to see

• gpt-4-0314 doen't exist it says (bug?)

Real adds

• Possibility to add multiple files at once
• When error because server is overloaded or so, wait x time and auto restart translation
• Auto save when finished (can be an option people need to enable) leep the file name as is and add the right language = file1.nl.srt
• Being able to scrull up without jumping down all the time
• ...

Describe the bug

Describe the problem you confuse

如题，不支持阅览上下文翻译 hmm

Describe the problem you confuse

还有希望添加批量选择字幕文件的功能

網頁版 一次放送30行 自動裝填 這個方法 好像 也不錯

Which method you use?

• gpt-4-0314
• gpt-3.5-turbo
• gpt-3.5-turbo-economy

Provide your subtitle file

Dungeons.And.Dragons.Honor.Among.Thieves.English-WWW.MY-SUBS.CO (1).srt.zip

Target language

Thai

Additional notes

None

Describe the bug

Throws a SyntaxError: Expected ',' or ']' after array element in JSON at position 144

To Reproduce

Normal use

I have translated the english subtitle into french and results are not correct
i mean the order

they are vtt but to upload here i made them txt

for example

00:13:37.270 --> 00:13:44.460
The fifth one is this is 200 face training
plus 14 epoch style training plus 20 more

translated as below

237
00:13:37,270 --> 00:13:44,460
La cinquième tentative inclut une formation de 200 époques sur le visage,
combinée à une formation sur le style de 14 époques, puis de 20 époques supplémentaires.
L'entraînement d'époques spécifié ici entraîne une déformation du visage et
endommage considérablement le style, comme vous pouvez le voir.

english.txt

french.txt

Describe the problem you confuse

There is no time interval, running too fast always prompts this

Describe the problem you confuse

测试了0.5版本 还是一行一行翻译，希望可以增加让它结合上下文的内容来翻译

Which method you use?

• gpt-4-0314
• gpt-3.5-turbo
• gpt-3.5-turbo-economy

Provide your subtitle file

Tousouchuu Great Mission.zip

Target language

Português Brasileiro

Additional notes

I noticed that sometimes after the \N the continuation of the text disappeared or something like \\\\\N appeared

Describe the bug

To Reproduce

我想是不是可以5行或者10行一起发送，
一是创造一个上下文环境，减少串行合并的情况；
二是降低频率，不太容易触发他的免费限制

一行字幕的字符数毕竟是有限的，我想10行应该也不大容易触碰到它的字符上限

Describe the bug

-- the day before
0.1 Open Subtitle Translator
0.2 choose English as target language
0.3 choose source file
0.4 translate
-> translation comes out ok
0.5 close Subtitle Translator

-- the day after
1.0 Open Subtitle Translator
1.1 choose source file
1.2 translate
-> instead of translating, it copies the subs

ADDITIONAL NOTES:
A. the first line is often left empty
B. the second line is the copy of the original language, translation starts from the third line

THANKS FOR THE AWESOME WORK!!!!

Sincere thanks!! From China,

Describe the problem you confuse

Hi, is it possible to export bilingual subtitles (including both the original and translated languages)?

Describe the problem you confuse

Can the function of saving project files be added to facilitate secondary adjustment of subtitle content?

coz i found its default installed in C disk which i dont really want it to be

翻译字幕最终结果总会有一些行未被生成中文字幕（只会合并两行英文），不知道是不是结合上下文的代码出了问题，望解决，非常感谢您的工具分享

Describe the problem you confuse

Hi, is it possible to switch the position of original subtitle and translated subtitle? The default setting seems that translated subtitle is on top of original subtitle, however, my use case prefer that original subtitle is on top. Thanks!

Describe the problem you confuse

如果是英文,会输出中文,否则相反
如果需要中英文双语字幕,如何解决,期望能解决这个问题

It happens a lot that during a translating the program halts due to unaivalability of the model:

Is it possible to include an option for auto resume?
That is, if this happens, the app will retry in 10 secs?

非付费用户一分钟只能发送三次请求，希望增加自动计时功能每分钟翻译三次而不是直接报错停止翻译

Describe the problem you confuse

Hi, Sir

Is it possible to save the translated SRT as bilingual? See attached picture for reference. If we want to make bilingual SRT file, this function is very useful if could provide the save option to save both original language and translated language into one single SRT file. Current, by default, only the translated target language could be saved as SRT and the original language was lost in the new saved SRT file. Thansk.

Describe the bug

When translating from Japanese to Simplified Chinese, the notes I added will be returned.

Describe the bug

Hello, your software is very good. I want to report a problem. If there is an error during the translation process and it stops, restarting the translation will start from the beginning, which is very unfriendly. I hope this can be optimized.

Describe the bug

This is a bug I encountered when using the api of openai with free quota, it seems that after I reached the speed limit, I didn't wait for enough cooling time and clicked translate again too fast to cause the bug, the right retranslate button disappeared and the problem of returning to the original text without returning the translation.

Wish I can just brew install --cask subtitle-translator-electron

Describe the problem you confuse

please share with me how to set up a proxy for this software?My operating system is Windows

Describe the problem you confuse

When using it, I found that a sentence needs 400tokens, so the subtitle of an episode needs about 10 tokens, which is too much consumption, can you reduce the consumption

I was using Wisper to transcribe the text, which caused a lot of break errors. A large number of repeated phrases appeared after using the translator. (A large number of sentence break errors resulted in the second half of the paragraph repeating the words of the first half)

Describe the problem you confuse

I am not sure if this only happens to me or not, I can't find how to save the subtitle, any hint, please?
I am using win ver.

After OpenAI releases the GPT-4 API, we should be able to integrate it as soon as possible. At that time, it may be possible to send the entire subtitle file in one request.