Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
My software needs to read a fixed-length handwritten number.
While I could use a general-purpose library like Tesseract, I am sure there is something smarter. Tesseract will probably misinterpret some of the 1 or 7 as I or l, whereas a software that expects only numbers would not.
Knowing that there are only numbers (American-English way of writing them), the algorithm could focus on 10 potential matches instead of hundreds of symbols.
Any experience OCRing handwritten number-only fields?
What open source library/software did you get the best results with?
From the FAQ of Tesseract:
How do I recognize only digits?
In 2.03 and above:
Use
TessBaseAPI::SetVariable("tessedit_char_whitelist", "0123456789");
before calling an Init function or put this in a text file called tessdata/configs/digits:
tessedit_char_whitelist 0123456789
and then your command line becomes:
tesseract image.tif outputbase nobatch digits
Warning: Until the old and new config variables get merged, you must have the nobatch parameter too.
But I think since it was designed for printed—not handwritten—text, accuracy might suffer even for digits only.
Related
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
Need to call some OCR function to extract characters from images. Wondering if anyone have good recommendations? Not necessarily to be open source and free.
thanks in advance,
Lin
When it come to OCR(optical character recognition) Tesseract is consider a better choice.
https://github.com/tesseract-ocr/tesseract
There is another library called microblink ocr. Code is available for ios, android and phone gap. Its accuracy is very good.
https://microblink.com/ocr
I have used GOCR a while ago, it's a command line tool, you can integrate it in your script if you are working server-side. I was satisfied with the result, it did a great job.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
I'm working on Sbire which use google speech-api. There is a some things which I don't like with this api.
First, this is not in streaming mode. It receive small audio files but there is no real pipes.
Secondly, it's not precise. Google does not record our voice to learn our accent. I think it's very important to have a good quality of recognition.
I'm looking for a free api or an open-source tool.
Actually there is a streaming mode, but it requires an API key. Here is an example of how to use it: http://mikepultz.com/2013/07/google-speech-api-full-duplex-php-version/
If you want something that learns, you'll need to create a neural network to handle the data. The best open source tool, however, is CMUSphinx or pocketSphinx(a mini version of CMUSphinx). Here is a link to that tool: http://sourceforge.net/p/cmusphinx/code/HEAD/tree/trunk/
The best part of CMU-Sphinx, however, is that you can set a dictionary of keywords. Instead of outputting gibberish, you can have it ONLY output those commands and finds the command closest to what it thinks you are saying. This make speech recognition incredibly accurate, but only work for a small set of words.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
Good evening,
I'm working on a project with Kinect and I need to perform a gender-recognition using the voice information caught by the microphone of the sensor.
The recognition could also be text-based, i.e. the sentence said by the user could always be the same. I "just" need the binary answer male/female.
I've checked the quite popular open-source Sphinx but I can't understand if it could be used for my needs.
Any idea of what I could use?
Thanks in advance.
For sake of completeness, finally I solved the recognition task without analyzing the voice input.
Gender can be detected with an acceptable precision also by learning and then classifying visual biometric traits. I used speech recognition only as "interface" to activate the demo.
There are several ways to achieve this. You can track pitch (lower pitch values will be male, otherwise female). Or try to build a GMM (Sphinx cannot do this, but HTK can), with one model for male, other for female and another to children.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
Legal department in my company wants a list of copyrights and licenses for all header files we use. They need this to verify we are using the right license and don't infringe any copyright.
So far I wrote a few simple bash lines to build everything with gcc -E, parse the output, locate the header files and grep the output for "copyright" and "license". This is very crude and error prone.
I am looking for an existing tool to do all of this, or at least parts:
Trace existing builds (like strace) and generate used headers instead of modifying the build system
Extract copyright holder and years from header files
Determine license per header file
Not looking for any legal advice here. Just looking for tools to help me easily analyze the code for our legal department.
You could contact folks like BlackDuck that essentially offer this analysis, by comparing your code base against a large set of open source code bases, whose licenses are already known. Any matches then yield the license for the matched code. They do this to answer precisely the question your lawyers are asking.
I have not used their product, nor have any association with them.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 8 years ago.
Improve this question
I'm looking for a Common Lisp implementation I ran across once, sometime in the past year or two. I only remember a few things, and I don't know how to search for it based on these facts, so maybe somebody here can help.
it was open-source, but wasn't one of the big ones (SBCL, CMUCL, MCL, etc.)
it was likely incomplete; it looked almost more like an exercise in writing the simplest possible self-hosted Common Lisp
the main webpage was plain black-on-white, and had 2 columns, where the left column was a link to the source file for a particular area of functionality (loop, format, clos, etc.), and the right column was a link to the tests for that functionality
the source files themselves were pretty-printed for the web, with syntax highlighting that looked kind of like an old Redhat Emacs default config: slate-gray background, etc.
Where can I find this Lisp implementation?
Thanks!
I don't know which one you are referring too, but you can find a list of Common Lisp Implementations here.
Is there any particular reason why this Lisp is grabbing your attention now?
Its hard to pin down, but open-source + minimalistic + incomplete sounds vaugely similar to Paul Graham's Arc programming language.