Is there a method to run this through the command line as simply as standard Whisper? I have a ton of issues with looking to do points by way of common Python code. I encounter way more VRAM difficulties, etcetera and it would not spit out all of different subtitle formats quickly.

Has any person performed that while? Once you have a great tuned model, Whisper is vastly more difficult to operate so far as I am able to tell.

I've employed subtitlecat to translate chinese subtitles posted listed here. However, when they have already got a translation they won't translate it and easily point out they already have a translation accessible . Also, it isn't a batch translation, but one at a time. Hope this assists.

Although some say Spikespen's Subs were not correct, they brought existence to JAV motion pictures for me, as someone who isn't going to have an understanding of the language and was amongst The explanations for me switching to Japanese AVs.​

@Imscully - you, together with @AgarabjeM, introduced me essentially the most gratification for my work on this thread = I've all the regard for the do the job and I'm glad you did not hand over!

three. To battle the last trouble I parsed my full map tree in two or three AI to inquire if it observed any duplicates which have this problem. Deepseek, Chatgpt could not take care of it, They simply invented replicate names. Claud.AI context window was way too compact (about 3000 actress names).

Some type souls have put up free portals both on Google Colab or to employ up their particular GPU cycles that they don't seem to be employing. If not, you will need to use Python or the command line.

Like another the latest AI designs, that is a big Neural Community (NN) trained on an enormous quantity of various knowledge. You will find lots of moving components to this, but what sets it aside from other transcriptions.

As far as I'm able to explain to, You can not just operate matters with The easy command line after that. You should commence worrying about the various parts of the procedure. Undecided if which includes adjusted or if it is going to adjust, nevertheless the documentation doesn't explain the method effectively.

Japanese is amongst the greater languages for this by typical transcription and translation metrics, but dirtiness and pronouns will be a Regular issue to work around.

That may be a lot of the benefit in functioning Whisper multiple times click here for every challenge. One of these will get an true translation in place of (speaking in international language).

It may not fully be a true translation considering that I do not know Japanese or Chinese. (Guess I ought to generally consist of this disclaimer!). Anyway delight in and allow me to know what you think that.

From there, open up it up in SubtitleEdit and discover the Strange or poor strains and watch the video clip to view what they need to be. SubtitleEdit appears like it had been built for modifying vehicle transcriptions on account of how every thing is laid out.

Many of these are thoroughly unedited, and The remainder are merely scarcely and have a bunch of mistakes. These are definitely not the exact same high quality as my Slave Shade subs.