Closed tvone closed 4 months ago
Love you tvone! This customized ver is fantastic!
I'm not quite clear on whether the custom model sends the context version or the non-context version? (Sorry I'm not very familiar with electron and js)
I saw this line in index-ea46f57b.js "["gpt-custom-model", "gpt-4", "gpt-3.5-turbo"].includes(g)". So I think this implies that custom-model will send context, right?
Another thing, I found that I can manually change the gpt-3.5 or gpt-4 model in the index-ea46f57b.js file inside the app.asar to gpt-4-turbo, or gpt-3.5-turbo, instead of always using the decrypted version, and then recompressing it back into the app.asar. Is this feasible?
Version 1.4.0 has been released. Please give it a try, and if you encounter any issues, feel free to raise an issue. https://github.com/gnehs/subtitle-translator-electron/releases/latest
Here is my customized version, while waiting for this repository to update https://github.com/tvone/subtitle-translator-electron/releases/tag/1.3.2