|
 |
Scanning +/- typing -> LWT ? | ||
Tags: Textbooks | ||
Share with:
| ||
 Language Learning Forum : Learning Techniques, Methods & Strategies |  | |
| schoenewaelder Diglot Senior Member Germany Joined 5372 days ago 759 posts - 1197 votes  Speaks: English*, French Studies: German, Spanish, Dutch | Message 1 of 4 16 April 2013 at 4:35pm | IP Logged |
|
I'd really quite like to OCR or type up my copy of Assimil's La Pratique du néerlandais. I know a lot of people like their little format, but I actually find them too small and not particularly easy to read. In this edition in particular, the "exercise" is really a second dialog, it is actually longer than the main dialogue, but the translations are in a smaller type, and the sentences all just run on, so it is just not matched up at all. And the reading environment of LWT, Readlang and Linq are much pleasanter
I must admit, I quite like clever tools like OCR stuff, and this will probably end up as a displacement activity rather than actual positive language learning. Pdf-Xchange and Cuneiform seem to give similar results on the OCR-ing, i.e., about 95% accurate, but that still leaves quite a bit of fiddling around, and to be honest, it would be almost as quick to type it up. Has anyone tried it, and if so, what sort of format did you try? I'm not sure if I really need the parallel texts format, as I'm happy to go through the dialogues multiple times in either language. Thanks. 1 person has voted this message useful | |
| twopossums Newbie United States Joined 4169 days ago 34 posts - 53 votes Speaks: English* | Message 2 of 4 16 April 2013 at 6:52pm | IP Logged |
|
What I've started doing recently is taking a screen grab of the pdf I want and them uploading it to Google Docs. Google Docs has free OCR built in. You can tell it what language you are uploading to help it out. Like yours its about 95% accurate so not perfect but it helps.
I didn't like the Assimil layout either. I make mine so I have the English sentence on top and the French below. That way I can cover up the French and try to translate. I'll spell out the words before revealing them. 2 persons have voted this message useful | |
| schoenewaelder Diglot Senior Member Germany Joined 5372 days ago 759 posts - 1197 votes Speaks: English*, French Studies: German, Spanish, Dutch | Message 3 of 4 17 April 2013 at 1:48pm | IP Logged |
|
Thanks.
It probably took just over half an hour to get one lesson in a basically usable form, but after that I was tiding it up, correcting "i" to "j" etc. for hours. But I was listening to the audio simultaneously, so it wasn't too much of a waste of time. The text document looks a bit boring at the moment, I'll have to try and think of a way to make it more visually appealing. However, with this new pdf viewer, I am now able to open the Assimil lesson accross three tabs, so I can set the zoom at the necessary level on each, and I don't keep messing up my place by scrolling up and down. Clicking between tabs cause no delay in rendering (as it does when paging up and down). Additionally, once it's been OCR-ed (the basic process just takes a couple of minutes per page, it doesn't change the appearance, but the viewer then knows where the lines are) I can highlight key or unknown words, which makes comparing the text and translation much easier. As well (or alternatively), using Audacity, I can just highlight the portion of audio relating to the page I'm on, and play it on loop. I should probably try and supress my perfectionist urges, and just accept this solution. 1 person has voted this message useful | |
| mrwarper Diglot Winner TAC 2012 Senior Member Spain forum_posts.asp?TID=Registered users can see my Skype Name Joined 5038 days ago 1493 posts - 2500 votes Speaks: Spanish*, EnglishC2 Studies: German, Russian, Japanese | Message 4 of 4 17 April 2013 at 6:58pm | IP Logged |
|
As OCR technology still has a long way to go, especially with complex text layouts, why don't you go for image-based scans? An OCRed text layer can be added later (and even replaced over time as OCR gets better) to help automated processing and use of software tools on your scan, but as a human being you work on what you *see* (final rendition of texts), so pictures at a high enough resolution work equally good for you, can be resized, zoomed in, etc. and are unaffected by OCR errors.
I can help with all things book scanning if you need any tips :) 3 persons have voted this message useful |
If you wish to post a reply to this topic you must first login. If you are not already registered you must first register
|  | 
Printable version
|
| Copyright 2024 FX Micheloud - All rights reserved No part of this website may be copied by any means without my written authorization. |
