Ramblings of an old Doc

 

My friends frequently post in foreign languages, and the meaning of the text often escapes me. It’s frustrating, because trying to copy the text can be time consuming or, in another alphabet. It’s especially frustrating when the text is part of an image.

For that reason, when I saw this in one of my favorite places (gHacks), I jumped on it…it’s small and free, so why not?

First off, it’ll appear like a fish on the right of the search box, and will be black if it works on the page or red if not. There are intermediate states though, as well: Translating some, but not all words, and not giving sense to idiomatic phrases…which you’ll have to guess. Also, it doesn’t detect all the fonts in an image equally. For example, it kind of works like the quick selection tool in Ps…contrast based. It can miss whole words for that reason, and can miss as text fonts which are impressionistic (dots) or ‘painterly’.

You don’t have to use the translation option in the extension, though. You can copy and paste it into Bing or Google Translate. They have their limitations as well, especially in highly idiomatic languages like French or Russian, and with slang or professional jargon.

For me, it’s a useful extension…you can always give it a try and remove it easily if it doesn’t serve your purpose.

It’s here if you want it: https://chrome.google.com/webstore/detail/copyfish-fresh-online-ocr/eenjdnjldapjajjofmldgmkjaienebbj

 

Source:

http://www.ghacks.net/2015/10/10/copyfish-for-chrome-copy-and-translate-text-from-media/?_m=3n%2e0038%2e1756%2ehj0ao01hy5%2e1tr3


Comments
on Oct 12, 2015

Hi,

i do use imtranslator for quick translations without personal content in compare mode.

Thats overfilled with ads  nod secure but better than nothing and working.

For OCR better buy and use  something like abby fine reader (above version 7)  if you are serious about it.

Sry but the only cheaper working more or less  alternative would be to search for a readiris  or textbridge(win3.x)  cdrom and prepare the pages with scantailor before passing through theses applications.

Yes ocr will be free open source software in next 5-10 years i belive , but it can`t beat the commercial ones recognition quality ( 95 % really for formatting and 98% for pure ocr  with  abby fine reader) at the moment.

 

However i`ll try it and see  if it works thx for the tip.