Is there a way to operate this by using the command line as basically as typical Whisper? I've a ton of issues with attempting to do items via standard Python code. I encounter far more VRAM concerns, etcetera and it would not spit out all of the various subtitle formats mechanically.
What exactly are the top softwares to develop the subtitle file And just how do they vary from one another? most straightforward one particular would be best.
one. Very first I wrote a script that checked each subtitle against the database of R18.dev. I believe script ran for around fifteen several hours to check about 20k subtitles. The trouble listed here is usually that actresses that debuted in the last 3 yr are only readily available within the database with their Kanji name within the r18 DB.
That is due to the fact there isn't any English subtitles for them. You will discover only a few individuals that translate JAV to English properly and just a few more who device translate and add. I only give Chinese subtitles for anybody asking for translations.
Our Local community has been around for quite some time and pride ourselves on featuring unbiased, crucial dialogue between men and women of all various backgrounds. We've been Functioning every single day to ensure our Local community is one of the better.
Start out-101 Eng Sub I went on a date With all the receptionist with the dentist’s Business office and it ended Along with the worst doable final result
⦁ Japanese is really a truly difficult language to have pronouns ideal. This really is something that can typically be corrected throughout the editing watch.
Would check here everyone know a Operating process to batch translate these information? The scripy deeplv4.py has stopped Functioning for me --I think a change in DeepL UI has induced it.
After some forwards and backwards with single sentences with ChatGPT I spotted Marie Konishi is alternating between calling the identical man or woman a naughty boy and filthy old gentleman, which designed Claude go nuts
two. Nearly anything that wasn't located through R18.dev database, there have been about 1600 subtitles still left from the main script, was then very first scraped vs Javguru and Next Javmost.
When you have any authorized difficulties remember to Get hold of the right media file owners or host websites or You can even Get hold of us. Serious about Advertising? Backlinks Trade? Get in touch with us: [email protected]
⦁ Though there are certainly going to be several web pages Later on that use Whisper transcriptions devoid of enhancing, a single editing move will probably tremendously Enhance the excellent and readability. It is not hard to capture poorly interpreted traces when executing a exam operate.
I actually respect these packs since OCRing subtitle has gotten more challenging in the latest months. Thanks much for uploading these packs!
If you want to know more about what other complications there will be In this particular pack you can read the workflow under.