Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Good (scriptable) OCR engine?
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Multimedia
View previous topic :: View next topic  
Author Message
avx
Veteran
Veteran


Joined: 21 Jun 2004
Posts: 1831

PostPosted: Fri Dec 30, 2011 12:04 pm    Post subject: Good (scriptable) OCR engine? Reply with quote

I'm in the need to OCR subtitles from DVDs, so that I can output them and use them as .srt. Currently I'm using avidemux and it's built-in OCR, but this makes huge problems for me, since it pretty much always uses "I" instead of "l" (big i instead of small L) or the other way round. While I can fix parts of this in an automated way, it's still a big task to fix the rest by hand - even with the support of a dictionary.

So my question is, which stand-alone OCR engine is good for this task and can ideally make use of a dictionary at runtime (german and english are most important here)? It should be somewhat scriptable and ideally doesn't need (too much) training, since the fonts on DVDs seem to be pretty much standard/comparable.

Thanks.
_________________
Want to thank me for something? Send me a nice postcard(ask per pm for my address), thank you!
Back to top
View user's profile Send private message
josehw
n00b
n00b


Joined: 02 Jan 2012
Posts: 13

PostPosted: Mon Jan 02, 2012 6:32 am    Post subject: Reply with quote

I'm sorry but I think no ocr engine can be accurate enough without being well-trained
Back to top
View user's profile Send private message
cord
Tux's lil' helper
Tux's lil' helper


Joined: 28 Apr 2007
Posts: 147
Location: Kharkov, UA

PostPosted: Mon Jan 02, 2012 2:37 pm    Post subject: Re: Good (scriptable) OCR engine? Reply with quote

avx wrote:
So my question is, which stand-alone OCR engine is good for this task and can ideally make use of a dictionary at runtime (german and english are most important here)?

Open source? - Seems to be no one. Only proprietary.
Back to top
View user's profile Send private message
avx
Veteran
Veteran


Joined: 21 Jun 2004
Posts: 1831

PostPosted: Mon Jan 02, 2012 3:54 pm    Post subject: Reply with quote

josehw wrote:
I'm sorry but I think no ocr engine can be accurate enough without being well-trained
Well, the problem is, avidemux shows me the images of I/l and I give it my correct answer, but even from there, it still doesn't work right most of the time.

chord wrote:
Open source? - Seems to be no one. Only proprietary.
I don't care about that, it just needs to work. I'll even pay for it, as long as there's a free trial around which doesn't want my data beforehand(ie, not like this: http://www.ocr4linux.com/en:trial).
_________________
Want to thank me for something? Send me a nice postcard(ask per pm for my address), thank you!
Back to top
View user's profile Send private message
cord
Tux's lil' helper
Tux's lil' helper


Joined: 28 Apr 2007
Posts: 147
Location: Kharkov, UA

PostPosted: Mon Jan 02, 2012 4:54 pm    Post subject: Reply with quote

avx wrote:
chord wrote:
Open source? - Seems to be no one. Only proprietary.
I don't care about that, it just needs to work. I'll even pay for it, as long as there's a free trial around which doesn't want my data beforehand(ie, not like this: http://www.ocr4linux.com/en:trial).

Fine Reader (not count as advertising). Here is how-to "Recognizing graphical PGS subtitles from Blu-Ray using Fine Reader" (also works with DVD, converts graphical subs to SRT). It's on russian. There must be it in english somewhere in the net (russian is my native, so I don't search for english). Good luck.

ps. If you can't understand, I can make translate for you at common steps :)
Back to top
View user's profile Send private message
avx
Veteran
Veteran


Joined: 21 Jun 2004
Posts: 1831

PostPosted: Mon Jan 02, 2012 8:35 pm    Post subject: Reply with quote

Looks like gibberish to me, but Google did the translation, so thanks. But from the pictures, this seems like Windows software and far from scriptable, so no good for me.

I'm trying cuneiform later tonight, will see if that's any good.
_________________
Want to thank me for something? Send me a nice postcard(ask per pm for my address), thank you!
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Multimedia All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum