Post your JAV subtitle files here - JAV Subtitle Repository (JSP)★NOT A SUB REQUEST THREAD★

Chuckie100 said:

HND-051 Reiko Kobayakawa A Thick Cum Fuck Sister Mother And A Blazing Son​

hnd051pl.jpg

As I indicated in my previous post, I liked this "reduced mosaic version" of this movie and subsequently decided to sub it and twist the storyline a bit! I used Whisper to create this sub and I attempted to clean it up a bit, and re-interpreted some of the meaningless/ "lewd-less" dialog. Again, I don't understand Japanese or Chinese so my re-interpretations might not be totally accurate but I try to match what is happening in the scene. Anyway, enjoy and let me know what you think.​

Click to expand...
Hey Chuckie,
I'm excited to see that you changed the brother-sister characters to mother/son.
It's EXACTLY the type of thing I like to do, so I doubly appreciate knowing that I have a brother in perverted spirit.
Thanks, as always, for your hard work and for sharing. Cheers
 
theydonotwantto said:
The whisper colab today failed with these message. I think some package are changed. Anyone know how to fix this?

Encoding audio...
Running VAD...

Using cache found in /root/.cache/torch/hub/snakers4_silero-vad_master

---------------------------------------------------------------------------

OSError Traceback (most recent call last)

<ipython-input-2-1b50d0fa71d6> in <cell line: 72>()
70
71 print("Running VAD...")
---> 72 model, utils = torch.hub.load(
73 repo_or_dir="snakers4/silero-vad", model="silero_vad", onnx=False
74 )

11 frames

/usr/lib/python3.10/ctypes/__init__.py in __init__(self, name, mode, handle, use_errno, use_last_error, winmode)
372
373 if handle is None:
--> 374 self._handle = _dlopen(self._name, mode)
375 else:
376 self._handle = handle

OSError: /usr/local/lib/python3.10/dist-packages/torchaudio/lib/libtorchaudio.so: undefined symbol: _ZN3c104cuda9SetDeviceEi
Click to expand...
Yes, Same. it seems the pip solver install incompatible versions of torchaudio and torch and the like. Hopefully somebody knows how to fix it.
 
mei2 said:
Give a try to this one: https://github.com/meizhong986/WhisperJAV

WhisperJAV


This one should work fine still. It uses faster-whisper with sileo vad --approx. 2x faster.
Note: this one requires a path to audio folder than to the audio itself. It subs all audios in the folder.
Click to expand...
Very Descriptive Name. Thanks to all the helpful people who created it. I've tried it out on SSIS-928. It kind of sucks.
 

Attachments

  • Like
Reactions: noolek and xsf27

Attachments

  • Like
Reactions: iko190806, noolek, Makkdom and 2 others
mei2 said:
Give a try to this one: https://github.com/meizhong986/WhisperJAV

WhisperJAV


This one should work fine still. It uses faster-whisper with sileo vad --approx. 2x faster.
Note: this one requires a path to audio folder than to the audio itself. It subs all audios in the folder.
Click to expand...
Does this one use large v2 model? Because if I am not mistaken, it is not indicated.

I will try to use this one if needed and maybe post some feedback regarding the results.

Thank you.
 
Prinsipe said:
Does this one use large v2 model? Because if I am not mistaken, it is not indicated.
Click to expand...
It does have the model as an option you can select in that colab so you are mistaken. It is the default.
 
  • Like
Reactions: mei2
ironfevers said:
I don't see VAD (Voice Activity Detection) in that version. It was a hassle to install https://colab.research.google.com/github/ANonEntity/WhisperWithVAD/blob/main/WhisperWithVAD.ipynb locally but it was worth it. There were so many things to install like CUDA, cudnn, (I think these two are already installed if you have used javplayer before), install https://pytorch.org, and

Python: 
pip install deepl srt ffmpeg-python
pip install git+https://github.com/openai/whisper.git

I have Python 3.9.13 and no compatibility issues. I can't say if Python 3.10 and newer have issues or not with packages like numpy, numba, etc. On my local pc, I changed the code so that it can queue up multiple audio paths. I can leave my pc on when I'm not home and whisper moves on to the next file. https://drive.google.com/file/d/1z1YX-YVgTZ2LoGDxmcK3BSYnZrWpSLYg/view?usp=share_link. I will explain the changes. Global variables for all the vad settings except audio_path is a parameter. Made a function called use_whisper and indented all the previous code into it. To queue up multiple audio, just copy paste use_whisper("your_path_to_audio_file") at the end (There is commented out code as an example). Also I added more words to the garbage list. Sounds like crackling and crying, and hallucinations such as, It's in my belly, and thank you for dinner gets deleted. To run, type python whisper-vad-function.py
Click to expand...
This has stopped working. The only advice I've seens is to only use ffmpeg-python, but that doesn't seem to fix it.
 
Besh said:
Whisper Large +VAD

MIAE-199

View attachment 3349012


-Besh
Click to expand...
At first I was disappointed because the only version of this Jav I could find was the one with the ubiquitous Chinese ads for the lottery and cam sex sites. Imagine my reaction when I discovered that your srt file was not only timed to this version, but it actually includes English translation for the lottery ad. After years of skipping through that damn thing, I now know what it is saying. Nice work!
 
  • Like
Reactions: Besh and mei2
quay2 said:
Did anyone manage to fix this? Seems to be a problem with Whisper being updated and dependencies not working anymore.
Click to expand...
It's the opposite actually, a requirement of the whisper + VAD process updated and whisper didn't so it forces an old version.

Cloning the whisper github and changing the triton line 16 in setup.py to require v 2.1.0 instead might fix it(and modify the VAD page to install the clone instead of the original whisper) or waiting for them to do it, assuming it doesn't break the code.

Code: 
requirements.append("triton==2.0.0")
Would become:
Code: 
requirements.append("triton==2.1.0")

I'd try it myself but I can't stay sitting for too long atm and other problems so would take a while until I can.

Edit: Or yeah, use the new JAV one, but I don't know if faster whisper is as good as the original.
 
  • Like
Reactions: quay2, xsf27 and mei2
SamKook said:
It's the opposite actually, a requirement of the whisper + VAD process updated and whisper didn't so it forces an old version.

Cloning the whisper github and changing the triton line 16 in setup.py to require v 2.1.0 instead might fix it(and modify the VAD page to install the clone instead of the original whisper) or waiting for them to do it, assuming it doesn't break the code.
Click to expand...
quay2 said:
Did anyone manage to fix this? Seems to be a problem with Whisper being updated and dependencies not working anymore.
Click to expand...

@SamKook , guys, I went ahead and fixed the Whisper Silero VAD for now. This should work. However, because I didn't have editor rights to the original one, I had to make a copy and to publish it from my github. I hope the orignal owner (aNoEntity) does not mind. I'm sure he doesn't.


WhisperWithVAD - maintenance release
It is saved here: https://github.com/meizhong986/WhisperJAV/tree/main/notebook

I have called the new version as a manitenance release, so it doesn't get mstaken from the original.

PS. if any issues let me know. I don't have much bandwidth but will try my best to fix it if this gets broken again.
PPS. the installation step takes a bit longer in this version because the cleanup, stay patient, it takes ca 3 minutes in total.
 
  • Like
Reactions: Arisaka Sana, Chuckie100, kanahana and 5 others
Another whisper sub pack, I translated and removed some rubbish text.

CMC-082
CMF-030
CMK-012
CMK-018
CMV-125
DBER-030
DBER-090
DBIK-001
DDAS-001
DDHG-002
DJUD-102
DPHD-012a
DPHD-012b
DPHD-014
DRNS-002
DXMG-021
JBD-145
KRU-073
RCT-684
RTP-008
SDMS-832
SDMT-245
SDMT-263
STARS-171
VICD-171
VSPD-024
 

Attachments

  • Like
  • Love
Reactions: shinjixx, noolek, Makkdom and 4 others
mei2 said:
@SamKook , guys, I went ahead and fixed the Whisper Silero VAD for now. This should work. However, because I didn't have editor rights to the original one, I had to make a copy and to publish it from my github. I hope the orignal owner (aNoEntity) does not mind. I'm sure he doesn't.


WhisperWithVAD - maintenance release
It is saved here: https://github.com/meizhong986/WhisperJAV/tree/main/notebook

I have called the new version as a manitenance release, so it doesn't get mstaken from the original.

PS. if any issues let me know. I don't have much bandwidth but will try my best to fix it if this gets broken again.
PPS. the installation step takes a bit longer in this version because the cleanup, stay patient, it takes ca 3 minutes in total.
Click to expand...

Thanks. It works. I replace the original setup part with your setup part and it works normally.
 
  • Like
Reactions: mei2
You must log in or register to reply here.
Top Bottom