recommend appropriate OCR program?

[BobLewiston]

Can anybody recommend an OCR (Optical Character Recognition) program that can do what I need it to do?

Here's the situation: Over a period of time, I figured out the chords to about 280 songs. Yeah, I know you can get the chord progressions to songs online, but they tend to be very inaccurate. I worked on each song until I was positive it was exactly right. (It made sense since I was getting paid to do it.) And I wrote down each song, by hand, one song per sheet of paper.

Now I want a permanent, neat record of all of these chord progressions, so I can print them out as requested.

So I scanned the song sheets I had written, with the idea of then translating them to text via an OCR program, and then formatting them in Word.

The problem is the OCR programs I tried. First I tried FreeOCR. It can’t handle handwriting at all - only typewriting. So then I tried SimpleOCR. This involved a lengthy exercise, the intent of which was for SimpleOCR to “learn” my handwriting. Towards this end, I wrote down by hand a newspaper article of about 500 words, and then scanned it and submitted it to SimpleOCR for analysis. This involved proofreading by me of SimpleOCR’s analysis.

But SimpleOCR’s performance was barely better than FreeOCR’s. Here is the problem, as I understand it:

Older OCR systems yielded limited accuracy because they simply compared scanned images of characters to stored bitmap images of known characters rendered in various specific fonts. More modern OCR systems achieve better accuracy because they can recognize characters based on the context of the word in which they appear (as with the Predictive Optical Word Recognition algorithm).

But in this case, this more modern approach is actually working against me. Apparently SimpleOCR thinks that anything that is submitted to it consists of sentences, and it’s trying to decipher each word based on having seen a sample, in the subject’s handwriting, of that word before.

But what I’m trying to translate into text isn’t a bunch of sentences composed of words. It’s chord progressions. Stuff like:

G6    Asus2    Asus2/#C    E    Bm7+11      
Am9    Bm7    CM7    D    Dsus4    D

SimpleOCR is completely failing to read this kind of thing correctly. It’s just spitting out gibberish.

Does anybody know what OCR program CAN handle this? Or failing that, can anyone recommend a more appropriate forum for this question?

[patio]

Reason will do this...however it's not cheap...

[BobLewiston]

Reason?  Do you mean ReadIris?  I can't find any "Reason" OCR online.

[Geek-9pm]

Did you read: CR software that recognizes handwritting

Some more information please.
What are your budget limits?
How much material do you have?
Does it have to be a computer only solution?
How long would it take a Human to type it out?
Can it be read by someone who knows English as a second language?
Can you provide some sample images?
Are there copyright issues?

[BobLewiston]

Geek-9pm:

Did you read: CR software that recognizes handwritting

Thanks for pointing this out. The URL referenced is

http://answers.google.com/answers/threadview/id/484756.html

Responding to some potentially helpful points made there:

Characttell?s SoftWriting can be downloaded free for a trial.

It is under $70 and only requires a standard TWAIN scanner (many inexpensive scanners are TWAIN compliant.)

http://www.charactell.com/SoftWriting.html

http://www.charactell.com/

http://www.charactell.com/SoftWritingDataSheet.pdf

I'll check that out.

If you have XP (even on a desktop) click on Start, Settings, Control Panel, Regional and Language, Languages, Details, and, under installed services, click ADD to see if handwriting recognition is installed already.

I've got Windows XP Pro, but handwriting recognition wasn't there.

If not, and you have Office XP installed, go to Add or Remove Programs, click on Office XP, then Change, then Add or remove features, then Next, then Office Shared Features, Alternative User
Input, Handwriting.

I've got Office XP, and I've followed these directions and installed Handwriting. Now how do I use this on a scanned sheet?

As for the connected, cursive handwriting, the chances of being able to convert that are slim.

No problem, none of this is in cursive.

Moving on to your questions:

What are your budget limits?

The cheaper, the better.

How much material do you have?

About 280 sheets.

Does it have to be a computer only solution?

I want as little labor as possible.

How long would it take a Human to type it out?

Maybe a full day. But because the chord progressions aren't composed of words, it would take much longer than you think to proofread to perfection.

Can it be read by someone who knows English as a second language?

It's chord progressions, like I showed (except where notes are added), so it's not a language at all.

Can you provide some sample images?

Let’s see if this works; I’ve never done this before. And just for the record, I transcribed whatever I was paid to; I didn’t choose the songs.

I can’t see the image in the preview; what did I do wrong? In between the img and /img demarcators I inserted the location on disk of the .jpg file. But maybe it’ll show when I actually post it (although I doubt it).

Are there copyright issues?

No - otherwise they would have long ago gone after all the people who post chord progressions on the net. Anyway, it's for my own use, not for publication.

[BobLewiston]

Update:

Softi Software's FreeOCR:
ABBYY FineReader 10 Pro:
They don't even try to process hand printed stuff.

Simple Software's SimpleOCR:
I.R.I.S. ReadIris Pro 11.0:
Too bad they tried - they accomplished nothing but wasting my time. I came up with a rather funny euphemism for their performance but unfortunately it's inappropriate for a public venue.

[rthompson80819]

I haven't used OCR software for several years and wasn't too happy with it then.

I can't remember which program I used, but I do remember it had a custom dictionary for words not in it's standard dictionary, just like in a word processor.

Maybe you need to look for something with that capability and you could put your chords in it.

[BC_Programmer]

I can’t see the image in the preview; what did I do wrong? In between the img and /img demarcators I inserted the location on disk of the .jpg file.

The image has to be uploaded to a server.

for example:

Code: [Select]

[img]http://bc-programming.com/avatar/mylogo.JPG[/img]
shows this:




You can use something like photobucket or another image hosting site, too.

[soybean]

The software that came with my Brother printer (actually, an All-in-One) will scan a document to text.  In other words, it performs OCR.  I haven't used the software to do this so I can't comment on how well it works.  Just thought I'd mention this in case you might have some OCR functionality already on your system that you weren't aware of.