Multiple lines all over the place. Timings wrong.
Perhaps AI generated. Needs human support.
English subtitles for [PURETABOO] To The Letter
Summary
- Created on: 2025-08-07 07:18:35
- Language:
English
- Ratings: Rate it!
- Duration: 50.22
- Comments: 6
Download
Filename:
puretaboo_to_the_letter__52377-20250807071835.zip
(8.9 KB)
Downloads:
Thanks:
1 "Thank You" received
Subtitles details
Name:
[PURETABOO] To The Letter (2025)
Duration:
50.22
Is only a draft:
No
Archive content:
Rebel Lynn - To The Letter - Pure Taboo.srt
Notes:
Preview:
Show/Hide subtitles preview ▼/▲
8
00:00:54,000 --> 00:01:00,000
Who are you?
9
00:01:00,000 --> 00:01:06,000
Who are you?
10
00:01:06,000 --> 00:01:12,000
Who are you?
11
00:01:12,000 --> 00:01:18,000
Who are you?
12
00:01:18,000 --> 00:01:24,000
Who are you?
13
00:01:48,000 --> 00:01:54,000
Who are you?
14
00:02:18,000 --> 00:02:24,000
Who are you?
15
00:02:24,000 --> 00:02:30,000
Who are you?
16
00:02:30,000 --> 00:02:36,000
Who are you?
17
00:02:36,000 --> 00:02:42,000
Who are you?
18
00:02:42,000 --> 00:02:48,000
Who are you?
19
00:02:48,000 --> 00:02:54,000
Who are you?
20
00:02:54,000 --> 00:03:00,000
Who are you?
21
00:03:00,000 --> 00:03:06,000
Who are you?
22
00:03:06,000 --> 00:03:12,000
Who are you?
23
00:03:12,000 --> 00:03:18,000
Who are you?
24
00:03:18,000 --> 00:03:24,000
Who are you?
25
00:03:24,000 --> 00:03:30,000
Who are you?
26
00:03:30,000 --> 00:03:36,000
Who are you?
27
00:03:36,000 --> 00:03:42,000
Who are you?
28
00:03:42,000 --> 00:03:48,000
Who are you?
00:00:54,000 --> 00:01:00,000
Who are you?
9
00:01:00,000 --> 00:01:06,000
Who are you?
10
00:01:06,000 --> 00:01:12,000
Who are you?
11
00:01:12,000 --> 00:01:18,000
Who are you?
12
00:01:18,000 --> 00:01:24,000
Who are you?
13
00:01:48,000 --> 00:01:54,000
Who are you?
14
00:02:18,000 --> 00:02:24,000
Who are you?
15
00:02:24,000 --> 00:02:30,000
Who are you?
16
00:02:30,000 --> 00:02:36,000
Who are you?
17
00:02:36,000 --> 00:02:42,000
Who are you?
18
00:02:42,000 --> 00:02:48,000
Who are you?
19
00:02:48,000 --> 00:02:54,000
Who are you?
20
00:02:54,000 --> 00:03:00,000
Who are you?
21
00:03:00,000 --> 00:03:06,000
Who are you?
22
00:03:06,000 --> 00:03:12,000
Who are you?
23
00:03:12,000 --> 00:03:18,000
Who are you?
24
00:03:18,000 --> 00:03:24,000
Who are you?
25
00:03:24,000 --> 00:03:30,000
Who are you?
26
00:03:30,000 --> 00:03:36,000
Who are you?
27
00:03:36,000 --> 00:03:42,000
Who are you?
28
00:03:42,000 --> 00:03:48,000
Who are you?
Ratings
You may like
Users comments
Hello NeoAV,
Maybe you're used to write subtitles somewhere else (so far we've never been able to see your work here). But if you do create something in order to share it for free, you should know it's a bit tiring to always get the same criticisms (ie form rather than content etc.)
I never commented it, because I'm sure you do it for the good of everyone, and you're even very positive sometimes!
But please, try to soften your tone and try to think "do this work deserve such a bad rating?"
Many place on internt ban unperfect works. It's not our choice here, and we are gentlemen (& ladies!)
Maybe you're used to write subtitles somewhere else (so far we've never been able to see your work here). But if you do create something in order to share it for free, you should know it's a bit tiring to always get the same criticisms (ie form rather than content etc.)
I never commented it, because I'm sure you do it for the good of everyone, and you're even very positive sometimes!
But please, try to soften your tone and try to think "do this work deserve such a bad rating?"
Many place on internt ban unperfect works. It's not our choice here, and we are gentlemen (& ladies!)
Just look at the subtitle preview, it's the Whisper hallucination. Requires more work.
So i think NeoAV is right.
But I'm working on a new Google Colab page, for using Faster-Whisper-XXL. I think the results will be better.
So i think NeoAV is right.
But I'm working on a new Google Colab page, for using Faster-Whisper-XXL. I think the results will be better.
My bad, sorry! I didn't notice.
When using Whisper, I often have the problem. It's not an ASR hallucination, this is more probably a segmentation problem. I never managed how to fix it, except trying again :(
When using Whisper, I often have the problem. It's not an ASR hallucination, this is more probably a segmentation problem. I never managed how to fix it, except trying again :(
one AI I tested asked "How many words per segment" I grew tired of so many broken lines, so I set it to 90. After that, just synch problems and a bit of missed dialogue, and of course, misunderstood mumbling/whispering. No matter what, work will be needed on any AI transcription, and if straight up AI, it should be marked as draft when posted, or make a new rating of AI Transcription. No difference, really.
• IA / Speech recognition (Whisper, VOSK, ...)
• I think they are very good / I did my best
• Sync precision is perfect or < 0.5s
• No translation (original language)