converting chinese text

[M.C.R.]

Hi,

 

is there anyone using software to convert an image with chinese characters into pinyin? I know it's possible, but I don't seem to get it right.

As a beginner I get a lot of homework that consists of translating texts of characters into Dutch. This takes a lot of time because of the many new characters. Instead of drawing and look-up every single character (I can't do copy paste). I would like to focus on memorizing characters, pinyin, pronunciation, and grammar.

I scanned some pages, saved them as TIFF images, tried to convert them into PDF and read the text with MS Office document imaging. But I can't add simplified chinese to the OCR of document imaging and my trial to convert to pdf has expired.

I think Abbyy finereader 11 and MDBG Chinese reader 7 are good, but with my studentbudget I can't afford it.

 

Can somebody please help me?

 

M.C.R.

 

 

 

[Yorin]

While something like this surely does exist, I will not work perfectly. Some Chinese characters have more than just one pronunciation, and it depends on context which pronunciation for a character is the right one in a given sentence.

 

More importantly, I question the way you want to learn Chinese. The point of the homework is obviously that you need to train how to read these characters. If you start OCR'ing whole texts even at the beginner level, you're not going to properly learn it. Of course, in the early stage such texts should only contain a very limited number of characters, and you should have been prepared on how to read them by your teacher or by whatever book you use in your classes. If your homework assignments always contain way too many characters you've never been shown before, then your teacher is not doing it right.

[imron]

Do you have a smartphone (iOS or Android)?  Pleco does OCR and the OCR module is not that expensive - only $10.

[Lu]

Generally, OCR doesn't really work for Chinese. Perhaps for some, clear characters, or in some cases, but you still want to be careful with this. There have been discussions on this forum before, you can search for it. One of my recent assignments was in fact going through a long, long text that included characters and finding & correcting the many, many instances where the OCR had gone wrong or given up. This took three people several months.

 

And yah, your homework sounds a bit weird. You're supposed to translate texts that contain a lot of words you've never seen before? If you can, perhaps you can discuss with your teacher and suggest using a different method. Perhaps first go over the new characters, together with the teacher, and then read the text.

[roddy]

What are the actual texts, they may be online? 

 

Seems to me that Chinese OCR is such a hassle you may as well find someone on elance or a similar site to type the stuff up for you. How much can a typist cost? 

[wibr]

I did a quick test with the current version of finereader for mac, the text was a scan (good quality but not great, around 300dpi, character size 45px to mostly 55px, relatively complicated layout with alternating traditional characters and English, roughly 1000 characters as sample set) and the accuracy for Chinese was around 99.5%, so not that bad I think.

You can download a trial version of most OCR systems, with the finereader trial you should be able to export the results for 100 x 3 pages within 30 days

[M.C.R.]

Hi,

 

many thanks for the replys.

I did sent a request for a trial of Abbyy finereader, but got an automatic e-mail and a link that doesn't work. Anyway, it would only be valid 30days, if it worked. I'm saving money to buy a smartphone. I don't have that, and I can see that the entire class who is using pleco on its smartphone, is making progress because it's easy and fast. Meanwhile I will continue translating the way I do now.

[James3]

Not sure if you have time for this approach or not, but a language exchange partner would probably be happy to help you out with this kind of thing. Help them practice their English first, and then when it comes to the "practicing Chinese" part of the session, send them some images and ask if they can type the Pinyin for them. Any time left over you can practice your spoken Chinese.