thank you for your tool to synchronize the funscript with remasters! I have used it with great success.
2 Likes
I was trying to remember the name of this plugin forever as I stupidly forgot to bookmark it. Iâm going to try to use it now for some Naughty America Remasters.
Edit: I see it logs the offsets to the console. Is there a flag that will save these results, or, even better, output to something like a CSV? Iâd like to capture this in an effort to write a simple tool to modify some cuepoints (aka bookmarks). XBVR and HereSphere (and probably DeoVR) use cuepoints, which can be synced from timestamp.trade, but often I run into problems where I have the cuepoints for the original scene but not the remaster. Since your program is already doing the hard work, and cuepoints can usually be edited in simple JSON (XBVR) and something similar with Stash, I am hoping it would be somewhat trivial to shift the cuepoints as well. If this isnât something that you would want to include in your app, I imagine that I could at least take the offset info and whip up a quick python script to query XBVR or Stash and generate a new set of cuepoints for the remaster.
This is a fantastic tool!
Right now, no, there isnât any option to save the offsets into a structured file (but it should be possible to parse the log file). It wouldnât be too hard to add an option like that thought. Iâll see what I can do when I have some time.
And, sure, I wouldnât mind adding the option to âtransformâ Xbvr/HereSphere if you find out the file format. From what I remember, Heresphere is using a binary format though so it might be a bit harder to do if itâs not documented.
1 Like
Yes to heresphere using binary, but it also imports from XBVR and Stash so you donât really have to mess with the binary ever if you are using either.
Hereâs an example using timestamp.trade and the XBVR (and I think Stash) format. Note that HereSphere itself is able to use cuepoints that have a duration, so advanced formats will show the beginning and end.
Title: Bye Bye Babysitter
Studio: WankzVR
Human-Readable Cuepoints: Cuepoints
XBVR Cuepoints: Import in JSON format
I realize that I can just pipe the output into a text file and then parse the text but I thought it would probably be better to dump the output via an easy-to-add option. Unfortunately I donât know any C# so I canât contribute a PR.
Here is an example of a different scene where the XBVR output that has multiple tracks for the cuepoints. The cuepoints also have start and end times. The tracks are used for things like adding position (sitting, standing, etc), adding which actresses are involved at which times (in MFF+ scenes), etc.
In the example below, track 0 is the ânormalâ cuepoint describing the action, and track 1 describes the camera position (important for VR).
{
"scene_id": "povr-2383253",
"cuepoints": [
{
"time_start": 783.629,
"time_end": 1013.016,
"track": 0,
"name": "cowgirl",
"rating": 0
},
{
"time_start": 1052.922,
"time_end": 1384.325,
"track": 0,
"name": "reverse cowgirl",
"rating": 0
},
{
"time_start": 1561.593,
"time_end": 1587.289,
"track": 0,
"name": "cumshot",
"rating": 0
},
{
"time_start": 233.189,
"time_end": 644.2,
"track": 1,
"name": "sitting",
"rating": 0
},
{
"time_start": 1482.288,
"time_end": 1561.593,
"track": 0,
"name": "Hand Job",
"rating": 0
},
{
"time_start": 319.243,
"time_end": 644.241,
"track": 0,
"name": "Blow Job",
"rating": 0
},
{
"time_start": 644.419,
"time_end": 779.917,
"track": 0,
"name": "Blow Job",
"rating": 0
},
{
"time_start": 1013.016,
"time_end": 1052.922,
"track": 0,
"name": "Blow Job",
"rating": 0
},
{
"time_start": 1384.325,
"time_end": 1482.288,
"track": 0,
"name": "Blow Job",
"rating": 0
},
{
"time_start": 644.4,
"time_end": 1611,
"track": 1,
"name": "lying",
"rating": 0
}
]
},
Hey @Zalunda Iâm trying to update a script to a remastered version, but in addition to cuts in the video, the framerate also changed between the old version I have and the remaster, and that seems to trip up FSTB. Is there a way to indicate the source framerate so itâs taken into account when using the audiosync.verifyfunscript action?
Or would my best option be to reencode the old version to the new framerate, fix the old funscript to the new framerate and then use FSTB from that to the remaster?
Usually, changing the framerate of the video (ex. 30 => 60fps) wouldnât change the audio track timing at all. So, itâs must be a âbad reencodingâ like VRPorn does (i.e. this).
Without knowing exactly how they âmessed upâ the reencoding, itâs hard to say whatâs the best approach. As you said, reencoding is an option but, in that case, I would suggest extracting the audio first and âexpanding/contractingâ only the audio stream. It should be a lot faster than reencoding the whole video (which is not used by FSTB anyway).
If the output of FSTB is giving you âstepâ offset (like this), you might be able to estimate the expansion/contraction that need to be applied.
Itâs a VRBangers remaster (Morning Muffins) so it very probably something similar to the VRPorn issue.
The remaster is 59.94006 fps and the previous version I had (from SLR) was 60fps.
The logs I get are:
From 00:00:00 to 00:00:04.933, 0 actions have been DROPPED
From 00:00:04.933 to 00:01:43.200, 155 actions have been MOVED by -00:00:04.933
From 00:01:43.200 to 00:01:43.208, 0 actions have been DROPPED
From 00:01:43.208 to 00:03:04.267, 248 actions have been MOVED by -00:00:04.942
From 00:03:04.267 to 00:03:41.750, 100 actions have been MOVED by -00:00:04.908
From 00:03:41.750 to 00:07:07.750, 724 actions have been MOVED by -00:00:04.817
From 00:07:07.750 to 00:07:07.783, 0 actions have been DROPPED
From 00:07:07.783 to 00:07:07.900, 0 actions have been MOVED by -00:00:04.850
From 00:07:07.900 to 00:09:31.800, 122 actions have been MOVED by -00:00:04.442
From 00:09:31.800 to 00:09:31.842, 0 actions have been DROPPED
From 00:09:31.842 to 00:10:31.492, 171 actions have been MOVED by -00:00:04.483
From 00:10:31.492 to 00:10:34.017, 9 actions have been MOVED by -00:00:04.425
From 00:10:34.017 to 00:10:34.083, 0 actions have been DROPPED
From 00:10:34.083 to 00:12:07.200, 238 actions have been MOVED by -00:00:04.492
From 00:12:07.200 to 00:12:08.075, 3 actions have been DROPPED
From 00:12:08.075 to 00:13:43.133, 359 actions have been MOVED by -00:00:05.367
From 00:13:43.133 to 00:15:10.275, 315 actions have been MOVED by -00:00:05.067
From 00:15:10.275 to 00:15:41.550, 89 actions have been MOVED by -00:00:05.042
From 00:15:41.550 to 00:15:42.225, 0 actions have been DROPPED
From 00:15:42.225 to 00:19:48.817, 0 actions have been MOVED by -00:00:05.717
From 00:19:48.817 to 00:19:57.517, 0 actions have been MOVED by -00:00:01.908
From 00:19:57.517 to 00:20:00.992, 0 actions have been DROPPED
From 00:20:00.992 to 00:22:36.083, 4 actions have been MOVED by -00:00:05.383
From 00:22:36.083 to 00:24:35.725, 419 actions have been DROPPED
From 00:24:35.725 to 00:26:00.867, 281 actions have been MOVED by -00:02:05.025
From 00:26:00.867 to 00:26:52.600, 147 actions have been DROPPED
From 00:26:52.600 to 00:27:14 , 27 actions have been MOVED by -00:02:56.758
From 00:27:14 to 00:30:00.258, 573 actions have been MOVED by -00:00:09.208
From 00:30:00.258 to 00:32:55.433, 640 actions have been MOVED by -00:00:08.858
But Iâm not too sure how you got the offset from the values in the log in your example.
Well in the end, I just shifted the script for the new frame rate with a script and then did shifts for the cuts manually in OFS 
@Zalunda Thank you again for linux compatible tools.
One question for you. After I resync for remastered, I usually delete the original video file but keep the orginal script. Do you see any benefit to making the .asig for the original video before I delete? Or just keep original video too?
Thanks!
Itâs up to you, but if you keep the original script, I suggest creating/keeping the .asig file. I wouldnât keep the original video since it would take a lot of space for nothing.
If âremaster #2â is available in the future, theoretically, it would be better to synchronize it with the original script and .asig instead of âremaster #1â's video/script. But unless you do that often (i.e. original => remaster #1 => remaster #2 => âŚ), it shouldnât make a big difference.
Iâve been trying to get the --novisual parameter to work when preparing video for OFS but I canât seem to get it in the correct place. It either breaks the operation or is ignored. On which line should I add it and am I correct that the line should read âânovisual=true ^â
You should be able to put it almost anywhere.
It should read --novisual (with 2 dashes). Note: Sometimes, on this board, the 2 dashes are replaced with a long dash.
If you place it alone on a line, ensure that the ^ is at the end of the line (i.e. no spaces after).
If it doesnât work, please tell me what error you are getting.
Have you encountered this before?
[13] INFO FunscriptToolbox.SubtitlesVerbs.VerbSubtitlesCreate - Video: [PurfviewWhisper][all] faster-whisper-xxl.exe: error: argument --output_format/-f: invalid choice: âC:\Users\Account\Downloads\FunscriptToolbox-1.3.5\FSTB-CreateSubtitles\Video_Backup\20250101212930-*.wavâ (choose from âjsonâ, âlrcâ, âtxtâ, âtextâ, âvttâ, âsrtâ, âtsvâ, âallâ)
[16] INFO FunscriptToolbox.SubtitlesVerbs.VerbSubtitlesCreate - Video: [PurfviewWhisper][all]
[15] INFO FunscriptToolbox.SubtitlesVerbs.VerbSubtitlesCreate - Video: [PurfviewWhisper][all]
[5] ERROR FunscriptToolbox.SubtitlesVerbs.VerbSubtitlesCreate - Video: An error occured while transcribing âfullâ:
Could not find file âC:\Users\Account\Downloads\FunscriptToolbox-1.3.5\FSTB-CreateSubtitles\Video_Backup\20250101212930-full-all.jsonâ.
Iâve tried installing large-v2 and large-v3.
Inserted the preconfig file, tried several videos as well. Any ideas?
The latest version of SubtitleEditor seems to install âfaster-whisper-xxl.exeâ instead of âwhisper-faster.exeâ. Iâll have to do some tests to see whatâs going on.
You should download the new version : Release 1.3.6 ¡ Zalunda/FunscriptToolbox ¡ GitHub
I had to change the order of the parameters sent to âfaster-whisper-xxl.exeâ. For some reason, --output_format json cannot be placed before the files list, but it works when I put it earlier on the command line.
Also, @MakeItFun, I finally took the time to add a --dumpoffsets parameter for the AudioSync.CreateFunscript command (and improved the output of that command as well).
1 Like
Hell yeah! Thanks dude. I am terrible at gui stuff but Iâm tempted to create some kind of basic controls to make this a little easier to use.
Thanks for your work! Any plans to have this updated to .NET 8/9 so it runs natively on Mac and Linux?
Thanks but I donât have immediate plans to switch to .NET 8/9. Maybe later, if I can get rid of the C code and Windows-specific UI part.
1 Like