What is Digg?Sponsored by Double Your Dating
The "How To Meet Women" Quiz view!
doubleyourdating.com - Three attractive women are standing a few feet away from you at a bar. What do you do? Find out...
25 Comments
- inactive, on 10/10/2007, -0/+11A well written article not the usual 3 paragraphs of obvious *****.
- Homunculiheaded, on 10/10/2007, -0/+10"Charles Wallace knows what a tesseract is"
- inactive, on 10/10/2007, -0/+7Yeah, too bad those softwares aren't open-source so that anyone could create additional functionality for them!
*shakes fist* - fkr3, on 10/10/2007, -1/+7I can't say I've ever seen someone tout no OCR as a reason not to use linux....
- lynkdead, on 10/10/2007, -1/+7Not nearly a cool enough project considering its name.
- neffy, on 10/10/2007, -0/+5Buried for getting me excited about intergalactic/interdimensional travel.
- Rhin, on 10/10/2007, -0/+5FreeOCR, a free OCR program for windows which uses the Tesseract engine:
http://softi.co.uk/freeocr.htm - schestowitz, on 10/10/2007, -4/+8A little old. :-) Here is the latest great development (albeit it depends on one's perspective):
Abiword gets experimental OCR scanning support
,----[ Quote ]
| An experimental Abiword plugin built on Ocropus and GEGL brings integrated
| OCR scanning support to the popular open source word processing program. The
| Abiword plugin is being developed by GNOME-scan creator Etienne Bersac.
`----
http://arstechnica.com/journals/linux.ars/2007/08/06/abiword-gets-experimental-ocr-scanning-support
Also of relevance:
Linux OCR: A review of free optical character recognition software
http://groundstate.ca/ocr
open source document analysis and OCR system
http://code.google.com/p/ocropus/
It is very important to squash the myth that GNU/Linux does not do OCR (and voice recognition). Many people use that out-of-date FUD to discourage adoption of Free software. - thinkingserious, on 10/10/2007, -1/+4What resources do you have for Linux voice recognition?
- indicas, on 10/10/2007, -0/+3i can tell it you it makes the easiest captchas useless, but on anything advanced it sucks.
- subliminalurge, on 10/10/2007, -0/+2Last time I looked into it there were several open-source options. All of them involved very convoluted installation procedures, including a list of dependencies a mile and a half long, and once finally installed and running, didn't perform very well.
That's been several years ago, though. I'm hoping someone will chime in and tell me the situation has improved since then.
At one time IBM also made ViaVoice available for free on Linux, which was easy to install and did a great job, but in recent times I've been unable to find a download source for that (or, indeed, any evidence that it ever existed on their website).
It would also be great to see some good open-source speech synthesis software become available. Might have to spend some time today checking into the current state of those projects. - units, on 10/10/2007, -0/+2Still, it's easily the best-performing f/oss OCR engine, which is of fairly high importance to academia; I started using this for machine print downstream processing in machine print zones at my job last summer.
- ewang, on 10/10/2007, -0/+2Do any, or all, of these work with character sets other than roman? IOW, could this be used to take an image from a japanese blog (for example) and put it in a format that could be submitted to Google Translate?
- rcgolf, on 10/10/2007, -0/+2Wonder if or how this may affect "captcha".........hmmmm
- jopsen, on 10/10/2007, -0/+2http://en.wikipedia.org/wiki/List_of_speech_recognition_projects
- krunchyfrog, on 10/10/2007, -0/+1Sweet, nonetheless
- Dankoozy, on 10/10/2007, -1/+2i tried this before and it wouldn't compile. maybe i'll have another go off it
- ApeInago, on 01/13/2008, -0/+1a little window can pop up at any time on the interface and will recognize anything you write into it and translate it to text. it has a bunch of nifty little productivity featurese too.
its like a fancy keyboard. - philpots, on 10/10/2007, -0/+1Handwriting recognition? -- I have many 19th Century plays (1 to 5 acts) and very difficult to read. One idea is to put all the groupings of words in list format and then eyeball and edit the obvious groups, then reassemble the script and try to made sense of the words that were not edited with in context of the edited words. This is from a end users perspective. There must be millions of handwritten documents out there to process.
- neodorian, on 10/10/2007, -0/+1ya rly. Where are the hypercubes?!
- ApeInago, on 10/10/2007, -0/+1I'd love to see a windows tablet pc edition equvalent functionality in linux...
- lucasmaximus, on 10/10/2007, -0/+0I've been using Tesseract OCR with a C# program I have made to batch OCR hundreds of documents a server we run. Considering it costs nothing, I am very impressed with the accuracy. It is far superior to GOCR which need the image to have the grey scale adjusted before anything can be done.
- Iam9376, on 10/10/2007, -0/+0Can you describe what your talking about? I've been curious about purchasing a tablet and developing a platform for Linux on Tablets , any feature recommendations would be helpful
- cozb, on 10/10/2007, -1/+0is there a open source OCR program that understands layouts? Something like does a good job like Abby FineReader?
- code2joy, on 10/10/2007, -7/+1open source software lacking features... that's fairly common.
© Digg Inc. 2009
— Content created and posted by Digg users is