Discussion of Indic language support in Tesseract OCR<br><br><div class="gmail_quote">---------- Forwarded message ----------<br>From: <b class="gmail_sendername">Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com">basilbthoppil@gmail.com</a>><br>
Date: 2008/6/12<br>Subject: dddd script/indic script support for TESSERACT<br>To: <a href="mailto:basilbthoppil@gmail.com">basilbthoppil@gmail.com</a><br><br><br><br><br><div class="gmail_quote">---------- Forwarded message ----------<br>
From: <b class="gmail_sendername">Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>Date: 2008/6/12<br>
Subject: Fwd: script/indic script support for TESSERACT<br>To: <a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a><br><br><br><br><br><div class="gmail_quote"><span style="font-size: large; font-weight: bold;">Forwarded conversation</span><br>
Subject: <b class="gmail_sendername">script/indic script support for TESSERACT</b><br>------------------------<br>
<br><span><font color="#000000">From: <b>Basil B Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>Date: Sun, Mar 9, 2008 at 8:18 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br>
</font><br><br></span><div>Hello everyone,<br><br> <br>I am planning to give indic language support for Tesseract OCR.<br><br>Is it possible to do complex script/indic script with the latest version of Tesseract.?<br>
<br>regards,<br>
<font color="#888888">basil<br>
</font></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>Date: Sun, Mar 9, 2008 at 10:52 PM<br>
To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span>Yes, it is possible.<div><div><div></div></div>
<br></div>
----------<br><span><font color="#000000">From: <b>Basil B Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>Date: Sun, Mar 9, 2008 at 11:05 PM<br>
To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Hello,<br>I am a newbie to Tesseract. As far as I know there is the problem of overlaping characters in Indic languages. How this is handled in Tesseract(How does the box is formed?).<br>
<br>Please mension more differences or problems we face when work with Tesseract.<br>
<br><br>regards <br><font color="#888888">basil</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Ray Smith</b> <<a href="mailto:theraysmith@gmail.com" target="_blank">theraysmith@gmail.com</a>><br>
Date: Sat, Apr 19, 2008 at 4:53 AM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Which Indic languages are you particularly interested in?<br>
Do you have any test data? (scanned images with the corresponding text.)<br>
I have done a lot of work with Kannada, but hit a bit of a stumbling block with its inherent ambiguity. Some of the other Indic languages might be easier.<br>
The Kannda work has led to a lot of changes that make it at least possible to consider the other Indic languages (and Chinese too).<br><font color="#888888">Ray.</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
Date: Sun, Apr 20, 2008 at 10:11 AM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Hi,<br>I am planning to do it for Malayalam. I dont know the case with kannada. The study is going on. There are some scanned images of some old books. Now I am planning to make some new scanned images.<br>
<br>There are some similar characters there in Malayalam which may cause ambiguity, but I hope it is solvable using DangAmbigous file.<br>
<br>regards,<br><font color="#888888">basil</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>
Date: Sun, Apr 20, 2008 at 12:23 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Please compare unicode for Malayalam with unicode for Kannada (<a href="http://unicode.org" target="_blank">unicode.org</a>).<br>
From the code chart of Malayalam, it is observed that Malayalam script has dependent vowels similar to Kannada script. If you are familiar with baraha software(windows platform), you can test Malayalam and Kannada scripts - by using transliteration from Malayalam to Kannada or Kannada to Malayalam This will prove that Malayalam can trained similar to Kannada easily<br>
</div>
please upload sample text of few character script (alphabets) of combination of consonants plus dependent vowels for testing..<div><div></div></div>----------<br><span><font color="#000000">From: <b>onkarswamy</b> <<a href="mailto:onkarswamy@gmail.com" target="_blank">onkarswamy@gmail.com</a>><br>
Date: Mon, Apr 21, 2008 at 5:36 PM<br>To: Basil B Thoppil <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br></font><br><br></span><div>hi sir i am also new to tesseract.i am also planing to develop ocr for<br>
kannada language(indic).please tell me how u proceed in this context.<br>
<br>
i have developed an on screen keyboard for kannada using BRHKND font<br>
<br>
i am currently in mysore ,sjce .<br>
<div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><div><br>
Date: Tue, Apr 22, 2008 at 11:49 AM<br></div>
To: onkarswamy <<a href="mailto:onkarswamy@gmail.com" target="_blank">onkarswamy@gmail.com</a>><br></font><br><br></span><div>Hi,<br>This is the scanned image of Malayalam.<div><div></div></div></div>
----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
Date: Fri, Apr 25, 2008 at 12:46 PM<div><br>To: onkarswamy <<a href="mailto:onkarswamy@gmail.com" target="_blank">onkarswamy@gmail.com</a>><br></div></font><br><br></span><br>Hi<div><br>
Hope you saw the document of Malayalam which I have attached.<br>
<br>regards,<br><font color="#888888">basil</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
Date: Fri, Apr 25, 2008 at 12:46 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div><div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
Date: Fri, Apr 25, 2008 at 12:49 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Hi,<br>Very sorry for the delay. It was wrongly sent to Mr <span>Onkarswamy.<br>
Please see the document and reply.<br><br>regards<br><font color="#888888">basil<br></font></span><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>onkarswamy angadi</b> <<a href="mailto:onkarswamy@gmail.com" target="_blank">onkarswamy@gmail.com</a>><br>
Date: Fri, Apr 25, 2008 at 3:16 PM<br>To: Basil Brassily Thoppil <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br></font><br><br></span><div>ya sir ,<br>what is my next step regarding the image?can u give some literature(or links) to read and get into next step.<br>
i am in mysore.where r u residing?<div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>
Date: Mon, Apr 28, 2008 at 12:11 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Hi Basil,<br>I am using MSwindows. Problem is I could not locte Malayalam fonts nor in Character Map in XP, as such I could not perform test nor send you a sample datafiles of Malayalam. I shall try again. <br>
<div><div></div></div><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>
Date: 2008/4/28<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a>, Basil Brassily Thoppil <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
</font><br>
<br></span><div>Hi Basil,<br>I have suceeded 4 lang. datafiles for tesseract-ocr as experiment. - vide attached files (1) 4lang.bmp(2)4lan.box(3)output text viz. 4lang.txt.<br>please compare output text viz.4lang.txt with 4lang.bmp file and 80% appears to be correct for malayam whereas 100% correct for Hindi and Kannada-80%.<br>
Since I don't know Malayalam script, please check and indicate number of mistakes occured.<div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>
Date: Mon, Apr 28, 2008 at 9:30 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>It is proved that indic langs can be trained in tesseract-ocr easily. Ray still working on to improve tesseract-ocr for Kannada. Once completed, I trust that all other Indian langs. can easily be trained in tesseract-ocr in the next version of tesseract-ocr. At present I have tested tesseract-ocr in MSwindows platform.<div>
<div></div></div></div>----------<br><span><font color="#000000">From: <b>dikar</b> <<a href="mailto:zhuhengguo@gmail.com" target="_blank">zhuhengguo@gmail.com</a>><br>Date: Thu, May 1, 2008 at 12:15 PM<br>
To: tesseract-ocr <<a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a>><br></font><br><br></span><div><br>
hi 74yrs old,<br>
you do a wonderful work!<br>
how do you finished that?<br>
I always get a error like "<br>
read_variables_file:variable not found: textord_no_rejectsTesseract<br>
Open Source OCR Engine<br>
Image has 24 bits per pixel and size (110,41)<br>
Resolution=96<br>
APPLY_BOXES: FATALITY - 0 labelled samples of "1" - target is 1<br>
APPLY_BOXES: FATALITY - 0 labelled samples of "0" - target is 2<br>
APPLY_BOXES:<br>
Boxes read from boxfile: 3<br>
Initially labelled blobs: 0 in 0 rows<br>
Box failures detected: 3<br>
Duped blobs for rebalance: 0<br>
"1" has fewest samples: 0<br>
Total unlabelled words: 0<br>
Final labelled words: 0<br>
Generating training data<br>
Generated training data for 0 blobs<br>
"<br>
I'm crazy almost.<br></div><div>
> 4lang.bmp<br>
> 1899KDownload<br>
><br>
> 4lang.box<br>
> 1KDownload<br>
><br>
> [4lang.txt]ka kA ki kI ku kU kIkU kAki ke kE kaki ko kEके kaಕೂ kakA kaki<br>
> ಕ ಕಾ ಕಿ ಕೀ ಕೠಕೂ ಕೃ ಕೄ ಕೆ ಕೇ ಕೈ ಕೊ ಕೊಕಾ ಕೌ ಕಂ ಕಃ<br>
> कं का कि की कॠकू कृ कॄ के कॆ कै को कॊ कौ कं कः<br>
> ക കാ കി കീ കൠകൂ കൃ കകൠಕಾക കേ കെക ಕಾകാ ಕಾകാ ಕಾകಕಾ കം കഃ<br>
><br>
> <a href="http://4lang.tr" target="_blank">4lang.tr</a><br>
> 182KDownload<br>
<div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>Date: Thu, May 1, 2008 at 1:27 PM<br>
To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Dikar,<br>Which language you are trying?will you forward image with text for perusal.<div>
<div></div></div></div>----------<br>
<span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>Date: Thu, May 1, 2008 at 3:34 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br>
</font><br><br></span><div>Dikar,<br>Please forward your trained material like image, box, text, tesseract.log files etc- which you find error. I shall try to locate where mistake happens and feedback to you. <br>
It may also indicate which operating system( like ubuntu or MSwindows etc) in which you tried;<br>
<font color="#888888">
-- <br><br></font><div class="gmail_quote"><div><div></div></div></div><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Zhu Hengguo</b> <<a href="mailto:zhuhengguo@gmail.com" target="_blank">zhuhengguo@gmail.com</a>><br>
Date: Thu, May 1, 2008 at 9:41 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div><br>I'm training some digital with handwriting.<br>
<br>when I use " tesseract fontfile.tif junk nobatch box.train" to make .tr file.<br>
<br>Program always metioned need <br>• tessdata/eng.freq-dawg <br>• tessdata/eng.word-dawg <br>
• tessdata/eng.user-words <br>• tessdata/eng.inttemp <br>• tessdata/eng.normproto <br>• tessdata/eng.pffmtable <br>• tessdata/eng.unicharset <br>• tessdata/eng.DangAmbigs <br><br>these files.<br>But before tr files have finished , inttemp ,pffmtable,normproto files have not prepared yet.<br>
When copy some examples and rename it as eng.inttemp , eng.normproto , eng.pffmtable; it report FATALITY occurs.<br>If use cnTraining.exe to make a null file in advance,it report error also.<br>So ,I'm failed to create .tr files<br>
<br>I just want to have a try on it ,so the box file and tif file is very simple .<br>like that "1","0"<div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
Date: Fri, May 2, 2008 at 12:08 AM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div><br>Hi 74 years old,<br><br>Very happy to see this much enthusiasm from your side for Indic languages. Thank you, thank you, very much. <br>
I also experimented with Malayalam and succeded, and I did in Gnu/Linux platform.<br>
I will attach the files in the next mail.<br><br><br>Thank you very much...<br><font color="#888888">Basil</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
Date: Fri, May 2, 2008 at 12:13 AM<br>To: 74yrs old <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br></font><br><br></span><div><br>Hi 74yrs old,<br><br>Will you please send me some training images for Hindi, that you used for training.!<br>
I have some doubts.<br>regards<br><font color="#888888">basil</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>
Date: Fri, May 2, 2008 at 1:37 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Hi Basil,<br>Glad to note you have succeeded Malayalam - congratulations. In fact I don't know Malayalam. In MSwindow - by using baraha software, I translated kannada to Malayalam and other languages and trained to prove that indic can be trained in tesseract-ocr easily. <br>
<br>In fact some of relevant source codes in tesseract-ocr have been modified by Ray, taking consideration of complexity in Kannada Script like combination of consonants plus dependent vowels, number of characters etc. I cannot explain since not I am not programmer nor developer. <br>
<br>All credit should go to Mr.Ray, owner of Tesseract-ocr, who has taken very much interest in developing Kannada project - which is most complex script among the Indic including other world languages which has combination of consonants plus dependent vowels etc. Still research is going on to perfect the relevant source code by Ray. so that all world languages can be trained easily in tesseract-ocr to the extent possible.<br>
<br>You can try out in MSplatform also where baraha software (<a href="http://www.baraha.com" target="_blank">www.baraha.com</a>) is available to train any other indic languages (using transliteration) easily.<br>With best Wishes.<div>
<div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>Date: Fri, May 2, 2008 at 2:33 PM<br>
To: Basil Brassily Thoppil <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br></font><br><br></span><div>Hi,<br>Please see attached file "4lang.bmp" forwarded to you on April, 28 - in which contains images of Hindi, Malayalam, Kannada etc What is your doubt?<br>
-sriranga(75yrsold)<div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>
Date: Fri, May 2, 2008 at 4:24 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Dikar,<br>Please find attached files. Output( hand.txt) agree with the smaple handwriting image(a.bmp). suceeded in handwriting test.<br>
<br><br><div class="gmail_quote"><div><div></div></div></div><div><div></div></div><br></div>1000<br>
<br>
<br>----------<br><span><font color="#000000">From: <b>Zhu Hengguo</b> <<a href="mailto:zhuhengguo@gmail.com" target="_blank">zhuhengguo@gmail.com</a>><br>Date: Fri, May 2, 2008 at 5:13 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br>
</font><br><br></span><div>Wow,magic! thanks a lot!<br><br></div><div class="gmail_quote">2008/5/2 74yrs old <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>>:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div></div></div>1000<br>
<br>
<br></blockquote></div><div><div></div></div>----------<br><span><font color="#000000">From: <b>Zhu Hengguo</b> <<a href="mailto:zhuhengguo@gmail.com" target="_blank">zhuhengguo@gmail.com</a>><br>
Date: Fri, May 2, 2008 at 7:48 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>hi 74yrs old,<br>What I meet is strange, but it works.<br>
The reason is :<br></div>
1 when you can do three things ,it maybe can work<div><br> 1.1 change .tif file to .bmp file (why, I don't know)<br>
1.2 if tif file support 24bit color ,change it to 1bit color.<br> 1.3 add something in command like that:<br> tesseract fontfile.tif junk -l aba nobatch box.train <br> aba is a language which you want to training.<br>
Note,if it metioned "need unicharset,...",you can use other tools(unicharset_extractor) make it in advance.<br> Note,another strange thing is when build 2.03 version, many thing work good but wordlist2dawg use a large mount memory and not stop.When I use old version,it works.<br>
<br>2 when <br> tesseract image.tif output -l lang<br> the output is empty.when I use image.bmp, it works!<br><br>thanks a lot!<br>And also thanks theraysmith and others, this ocr is a good ocr.<div><div></div>
</div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>Date: Fri, May 2, 2008 at 11:37 PM<br>
To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Zhu,<br><u>Reg:wordlist_word/freq_drawig in tesseract2-03</u>: Yes,there is problem as you reported - I have already brought to the kind notice of Ray. Wait for the patch. In fact, I have tested in tesseract 2.01 only.<br>
<br>Reg:1.1: tif was scanned at 96 dpi -as such blank. If you had scanned at 300 dpi, it would have work. With help of Paint brush I saved "a.tif" as "a.bmp file". <br><br>Reg. 1.2 Here not question of 24 colors,but question of dpi - morever tif should be uncompressed plus scanned at least 300dpi or more. <br>
<br>Reg 1.3 I have not added anything in the commandline. I followed wiki instructions for training. All generated datafiles should be prefixed three characters like aba or han etc.- for example generated datafile as "unicharset" - for which you have to prefix like "aba.unicharset" or han.unicharset etc. <br>
<br>Reg:2 Yes because tif was scanned at 96 dpi and <br>not 300- dpi. as such output was <a href="http://empty.in" target="_blank">empty.in</a> your case.<br><br>Please note that for handwriting - it requires more training for single character due to variation of shape,size etc while writing. . I feel steady/uniform handwritings may succeed <br>
<div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>Date: Sat, May 3, 2008 at 12:38 PM<br>
To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Hi 78yrs old,<br>I havent done the work completely. It was just for experimentation purpose. I trained some single characters only. But there are a lot of combinations of letters and many rendering problems in Malayalam. We have to consider that also.<br>
<br>for eg.<br><br>ഗ + ് + ന becomes ഗ്ന I am just coming to that part. Given for your research purpose.<br><br><br>regards<br><font color="#888888">basil.</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>Basil Brassily Thoppil</b> <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br>
Date: Sat, May 3, 2008 at 12:47 PM<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Hi,<br>about Hindi,<br>please see the attached image of a Hindi document.<br>
There is a line connecting the letters of each word. Then how does tesseract idenntify a single letter from a document? Is there any special feature .?<br>
regards<br><font color="#888888">basil</font><div><div></div></div></div>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>
Date: 2008/5/3<br>To: <a href="mailto:tesseract-ocr@googlegroups.com" target="_blank">tesseract-ocr@googlegroups.com</a><br></font><br><br></span><div>Basil,<br>As an example please attached file - which is self explanatory about each character has combinations of consonants plus vowel. and such each character has to be trained -.read with wiki about training -.and generate Malayalam datafiles (similar to English datafiles) etc. <br>
<br>I can understand your example - which have already been considered by Ray, for the latest version of tesseract-2.03. As such there may not any problem with Malayam which is similar to Kannada.<br><br>Trust you have hands-on experience with the English (phototest.tif) and also with english words or text you might have experimented. .<div>
<div></div></div><br><br>
<br></div>
അ ആ ഇ ഈ ഉ ഊ ഋ ൠ എ ഏ ഐ ഒ ഓ ഔ അം അഃ<br>
ക കാ കി കീ കു കൂ കൃ കൠ കെ കേ കൈ കൊ കോ കൌ കം കഃ<br>
ഗ ഗാ ഗി ഗീ ഗു ഗൂ ഗൃ ഗൠ ഗെ ഗേ ഗൈ ഗൊ ഗോ ഗൌ ഗം ഗഃ<br>
ഖ ഖാ ഖി ഖീ ഖു ഖൂ ഖൃ ഖൠ ഖെ ഖേ ഖൈ ഖൊ ഖോ ഖൌ ഖം ഖഃ<br>
ഘ ഘാ ഘി ഘീ ഘു ഘൂ ഘൃ ഘൠ ഘെ ഘേ ഘൈ ഘൊ ഘോ ഘൌ ഘം ഘഃ<br>
ങ ങാ ങി ങീ ങു ങൂ ങൃ ങൠ ങെ ങേ ങൈ ങൊ ങോ ങൌ ങം ങഃ<br>
ച ചാ ചി ചീ ചു ചൂ ചൃ ചൠ ചെ ചേ ചൈ ചൊ ചോ ചൌ ചം ചഃ<br>
ഛ ഛാ ഛി ഛീ ഛു ഛൂ ഛൃ ഛൠ ഛെ ഛേ ഛൈ ഛൊ ഛോ ഛൌ ഛം ഛഃ<br>
ജ ജാ ജി ജീ ജു ജൂ ജൃ ജൠ ജെ ജേ ജൈ ജൊ ജോ ജൌ ജം ജഃ<br>
ഝ ഝാ ഝി ഝീ ഝു ഝൂ ഝൃ ഝൠ ഝെ ഝേ ഝൈ ഝൊ ഝോ ഝൌ ഝം ഝഃ<br>
൦ ൧ ൨൩ ൪൫ ൬ ൭ ൮ ൯ : ; < = > ? ॒ [ \ ] _ { | }<br>
~| || ! " # % & ' ( ) * + , - . /<br>
ಞ ಞಾ ಞಿ ಞೀ ಞು ಞೂ ಞೃ ಞೄ ಞೆ ಞೇ ಞೈ ಞೊ ಞೋ ಞೌ ಞಂ ಞಃ<br>
ಟ ಟಾ ಟಿ ಟೀ ಟು ಟೂ ಟೃ ಟೄ ಟಿ ಟೀ ಟೈ ಟೊ ಟೋ ಟೌ ಟಂ ಟಃ<br>
ಠ ಠಾ ಠಿ ಠೀ ಠು ಠೂ ಠೃ ಠೄ ಠೆ ಠೇ ಠೈ ಠೊ ಠೋ ಠೌ ಠಂ ಠಃ<br>
ಡ ಡಾ ಡಿ ಡೀ ಡು ಡೂ ಡೃ ಡೄ ಡೆ ಡೇ ಡೈ ಡೊ ಡೋ ಡೌ ಡಂ ಡಃ<br>
ಢ ಢಾ ಢಿ ಢೀ ಢು ಢೂ ಢೃ ಢೄ ಢೆ ಢೇ ಢೈ ಢೊ ಢೋ ಢೌ ಢಂ ಢಃ<br>
ಣ ಣಾ ಣಿ ಣೀ ಣು ಣೂ ಣೃ ಣೄ ಣೆ ಣೇ ಣೈ ಣೊ ಣೋ ಣೌ ಣಂ ಣಃ<br>
ತ ತಾ ತಿ ತೀ ತು ತೂ ತೃ ತೄ ತೆ ತೇ ತೈ ತೊ ತೋ ತೌ ತಂ ತಃ<br>
ಥ ಥಾ ಥಿ ಥೀ ಥು ಥೂ ಥೃ ಥೄ ಥೆ ಥೇ ಥೈ ಥೊ ಥೋ ಥೌ ಥಂ ಥಃ<br>
ದ ದಾ ದಿ ದೀ ದು ದೂ ದೃ ದೄ ದೆ ದೇ ದೈ ದೊ ದೋ ದೌ ದಂ ದಃ<br>
ಧ ಧಾ ಧಿ ಧೀ ಧು ಧೂ ಧೃ ಧೄ ಧೆ ಧೇ ಧೈ ಧೊ ಧೋ ಧೌ ಧಂ ಧಃ<br>
ನ ನಾ ನಿ ನೀ ನು ನೂ ನೃ ನೄ ನೆ ನೇ ನೈ ನೊ ನೋ ನೌ ನಂ ನಃ<br>
ಪ ಪಾ ಪಿ ಪೀ ಪು ಪೂ ಪೃ ಪೄ ಪೆ ಪೇ ಪೈ ಪೊ ಪೋ ಪೌ ಪಂ ಪಃ<br>
ಫ ಫಾ ಫಿ ಫೀ ಫು ಫೂ ಫೃ ಫೄ ಫೆ ಫೇ ಫೈ ಫೊ ಫೋ ಫೌ ಫಂ ಫಃ<br>
ಬ ಬಾ ಬಿ ಬೀ ಬು ಬೂ ಬೃ ಬೄ ಬೆ ಬೇ ಬೈ ಬೊ ಬೋ ಬೌ ಬಂ ಬಃ<br>
ಭ ಭಾ ಭಿ ಭೀ ಭು ಭೂ ಭೃ ಭೄ ಭೆ ಭೇ ಭೈ ಭೊ ಭೋ ಭೌ ಭಂ ಭಃ<br>
ಮ ಮಾ ಮಿ ಮೀ ಮು ಮೂ ಮೃ ಮೄ ಮೆ ಮೇ ಮೈ ಮೊ ಮೋ ಮೌ ಮಂ ಮಃ<br>
ಯ ಯಾ ಯಿ ಯೀ ಯು ಯೂ ಯೃ ಯೄ ಯೆ ಯೇ ಯೈ ಯೊ ಯೋ ಯೌ ಯಂ ಯಃ<br>
ರ ರಾ ರಿ ರೀ ರು ರೂ ರೃ ರೄ ರೆ ರೇ ರೈ ರೊ ರೋ ರೌ ರಂ ರಃ<br>
ಲ ಲಾ ಲಿ ಲೀ ಲು ಲೂ ಲೃ ಲೄ ಲೆ ಲೇ ಲೈ ಲೊ ಲೋ ಲೌ ಲಂ ಲಃ<br>
ಳ ಳಾ ಳಿ ಳೀ ಳು ಳೂ ಳೃ ಳೄ ಳೆ ಳೇ ಳೈ ಳೊ ಳೋ ಳೌ ಳಂ ಳಃ<br>
ವ ವಾ ವಿ ವೀ ವು ವೂ ವೃ ವೄ ವೆ ವೇ ವೈ ವೊ ವೋ ವೌ ವಂ ವಃ<br>
ಶ ಶಾ ಶಿ ಶೀ ಶು ಶೂ ಶೃ ಶೄ ಶೆ ಶೇ ಶೈ ಶೊ ಶೋ ಶೌ ಶಂ ಶಃ<br>
ಷ ಷಾ ಷಿ ಷೀ ಷು ಷೂ ಷೃ ಷೄ ಷೆ ಷೇ ಷೈ ಷೊ ಷೋ ಷೌ ಷಂ ಷಃ<br>
ಸ ಸಾ ಸಿ ಸೀ ಸು ಸೂ ಸೃ ಸೄ ಸೆ ಸೇ ಸೈ ಸೊ ಸೋ ಸೌ ಸಂ ಸಃ<br>
ಹ ಹಾ ಹಿ ಹೀ ಹು ಹೂ ಹೃ ಹೄ ಹೆ ಹೇ ಹೈ ಹೊ ಹೋ ಹೌ ಹಂ ಹಃ<br>
ಕ್ಕ ಖ್ಖ ಗ್ಗ ಘ್ಘ ಙ್ಙ ಚ್ಚ ಛ್ಛ ಜ್ಜ ಝ್ಝ ಞ್ಞ ಟ್ಟ ಠ್ಠ ಡ್ಡ ಢ್ಢ ಣ್ಣ ತ್ತ ಥ್ಥ ದ್ದ ಧ್ಧ ನ್ನ ಪ್ಪ ಫ್ಫ ಬ್ಬ ಭ್ಭ ಮ್ಮ<br>
ಯ್ಯ ರ್ರ ಲ್ಲ ಳ್ಳ ವ್ವ ಶ್ಶ ಷ್ಷ ಸ್ಸ ಹ್ಹ ಕ್ರ ಖ್ರ ಗ್ರ ಘ್ರ ಙ್ರ ಚ್ರ ಛ್ರ ಜ್ರ ಝ್ರ ಞ್ರ ಠ್ರ ಡ್ರ ಢ್ರ ಣ್ರ<br>
ತ್ರ ಥ್ರ ದ್ರ ಧ್ರ ನ್ರ ಪ್ರ ಫ್ರ ಬ್ರ ಭ್ರ ಮ್ರ ಯ್ರ ರ್ರ ಲ್ರ ಳ್ರ ವ್ರ ಶ್ರ ಷ್ರ ಸ್ರ ಹ್ರ ೦ ೧ ೨ ೩ ೪ ೫ ೬ ೭ ೮ ೯<br>
ಶ್ಟು ಷ್ಟು ಕ್ಟ್ ಗ್ಧ ದ್ಧ ದ್ಧೆ ಸ್ದೇ ದ್ಭು ಶ್ರೀ ಸ್ತ್ರೀ ಪ್ರ ಸ್ರ ಣ್ಯ ಧ್ಯ ವ್ಯಾ ಪ್ಯಾ ಲ್ಯ ತ್ಯ ತ್ತ್ಯನ್ಯೋ ಸ್ಯಾನ್ಯಃ ರ್ಣಾ ರ್ತೃ ರ್ಬ ತ್ತ್ವ ಶ್ವೇ ಶ್ವ ಸ್ವ ದ್ವಾ ದ್ವಾ ಸ್ವಾ ದ್ವ ಜ್ಞಾ ಜ್ಞ ಬ್ರ ಹ್ಮಾಂ ನ್ಮ ಹ್ಮ ಲ್ಮೀ ಸ್ಪಶ್ನ ಸ್ತ ಯ್ತು ಪ್ತ ಷ್ಟ್ರೀ ಸ್ಟೆ ಫ್ಲಾ ಕ್ಟ್ರಿ ಷ್ಠೆ ಶ್ಚಾ ತ್ಯ ಕ್ಪ್ಮ ಜ್ಞ ಷ್ಟ್ರ ಕ್ಷ್ಮಿ ಸ್ಥಿ ಷ್ಟ್ರಾ ವೈ ದ್ಯ್ಕ ನ್ಕೈ<br>
ಕ್ಷಂ ಕ್ಷ ಕ್ಲಾ ಥ್ಯಾಂ ಕ್ಸ್. ಹ್ಯಾಂ ಲ್ಮೇ ಲ್ಡ<br>
<br>
<br>
<br>
<br>
<br>
<br>
ಅ ಆ ಇ ಈ ಉ ಊ ಋ ೠ ಎ ಏ ಐ ಒ ಓ ಔ ಅಂ ಅಃ<br>
ಕ ಕಾ ಕಿ ಕೀ ಕು ಕೂ ಕೃ ಕೄ ಕೆ ಕೇ ಕೈ ಕೊ ಕೋ ಕೌ ಕಂ ಕಃ<br>
ಗ ಗಾ ಗಿ ಗೀ ಗು ಗೂ ಗೃ ಗೄ ಗೆ ಗೇ ಗೈ ಗೊ ಗೋ ಗೌ ಗಂ ಗಃ<br>
ಖ ಖಾ ಖಿ ಖೀ ಖು ಖೂ ಖೃ ಖೄ ಖೆ ಖೇ ಖೈ ಖೊ ಖೋ ಖೌ ಖಂ ಖಃ<br>
ಘ ಘಾ ಘಿ ಘೀ ಘು ಘೂ ಘೃ ಘೄ ಘೆ ಘೇ ಘೈ ಘೊ ಘೋ ಘೌ ಘಂ ಘಃ<br>
ಙ ಙಾ ಙಿ ಙೀ ಙು ಙೂ ಙೃ ಙೄ ಙೆ ಙೇ ಙೈ ಙೊ ಙೋ ಙೌ ಙಂ ಙಃ<br>
ಚ ಚಾ ಚಿ ಚೀ ಚು ಚೂ ಚೃ ಚೄ ಚೆ ಚೇ ಚೈ ಚೊ ಚೋ ಚೌ ಚಂ ಚಃ<br>
ಛ ಛಾ ಛಿ ಛೀ ಛು ಛೂ ಛೃ ಛೄ ಛೆ ಛೇ ಛೈ ಛೊ ಛೋ ಛೌ ಛಂ ಛಃ<br>
ಜ ಜಾ ಜಿ ಜೀ ಜು ಜೂ ಜೃ ಜೄ ಜೆ ಜೇ ಜೈ ಜೊ ಜೋ ಜೌ ಜಂ ಜಃ<br>
ಝ ಝಾ ಝಿ ಝೀ ಝು ಝೂ ಝೃ ಝೄ ಝೆ ಝೇ ಝೈ ಝೊ ಝೋ ಝೌ ಝಂ ಝಃ<br>
0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ [ \ ] ^ _ { | }<br>
~÷। ॥ ! " # % & ' ( ) * + , - . /<br>
ಞ ಞಾ ಞಿ ಞೀ ಞು ಞೂ ಞೃ ಞೄ ಞೆ ಞೇ ಞೈ ಞೊ ಞೋ ಞೌ ಞಂ ಞಃ<br>
ಟ ಟಾ ಟಿ ಟೀ ಟು ಟೂ ಟೃ ಟೄ ಟಿ ಟೀ ಟೈ ಟೊ ಟೋ ಟೌ ಟಂ ಟಃ<br>
ಠ ಠಾ ಠಿ ಠೀ ಠು ಠೂ ಠೃ ಠೄ ಠೆ ಠೇ ಠೈ ಠೊ ಠೋ ಠೌ ಠಂ ಠಃ<br>
ಡ ಡಾ ಡಿ ಡೀ ಡು ಡೂ ಡೃ ಡೄ ಡೆ ಡೇ ಡೈ ಡೊ ಡೋ ಡೌ ಡಂ ಡಃ<br>
ಢ ಢಾ ಢಿ ಢೀ ಢು ಢೂ ಢೃ ಢೄ ಢೆ ಢೇ ಢೈ ಢೊ ಢೋ ಢೌ ಢಂ ಢಃ<br>
ಣ ಣಾ ಣಿ ಣೀ ಣು ಣೂ ಣೃ ಣೄ ಣೆ ಣೇ ಣೈ ಣೊ ಣೋ ಣೌ ಣಂ ಣಃ<br>
ತ ತಾ ತಿ ತೀ ತು ತೂ ತೃ ತೄ ತೆ ತೇ ತೈ ತೊ ತೋ ತೌ ತಂ ತಃ<br>
ಥ ಥಾ ಥಿ ಥೀ ಥು ಥೂ ಥೃ ಥೄ ಥೆ ಥೇ ಥೈ ಥೊ ಥೋ ಥೌ ಥಂ ಥಃ<br>
ದ ದಾ ದಿ ದೀ ದು ದೂ ದೃ ದೄ ದೆ ದೇ ದೈ ದೊ ದೋ ದೌ ದಂ ದಃ<br>
ಧ ಧಾ ಧಿ ಧೀ ಧು ಧೂ ಧೃ ಧೄ ಧೆ ಧೇ ಧೈ ಧೊ ಧೋ ಧೌ ಧಂ ಧಃ<br>
ನ ನಾ ನಿ ನೀ ನು ನೂ ನೃ ನೄ ನೆ ನೇ ನೈ ನೊ ನೋ ನೌ ನಂ ನಃ<br>
ಪ ಪಾ ಪಿ ಪೀ ಪು ಪೂ ಪೃ ಪೄ ಪೆ ಪೇ ಪೈ ಪೊ ಪೋ ಪೌ ಪಂ ಪಃ<br>
ಫ ಫಾ ಫಿ ಫೀ ಫು ಫೂ ಫೃ ಫೄ ಫೆ ಫೇ ಫೈ ಫೊ ಫೋ ಫೌ ಫಂ ಫಃ<br>
ಬ ಬಾ ಬಿ ಬೀ ಬು ಬೂ ಬೃ ಬೄ ಬೆ ಬೇ ಬೈ ಬೊ ಬೋ ಬೌ ಬಂ ಬಃ<br>
ಭ ಭಾ ಭಿ ಭೀ ಭು ಭೂ ಭೃ ಭೄ ಭೆ ಭೇ ಭೈ ಭೊ ಭೋ ಭೌ ಭಂ ಭಃ<br>
ಮ ಮಾ ಮಿ ಮೀ ಮು ಮೂ ಮೃ ಮೄ ಮೆ ಮೇ ಮೈ ಮೊ ಮೋ ಮೌ ಮಂ ಮಃ<br>
ಯ ಯಾ ಯಿ ಯೀ ಯು ಯೂ ಯೃ ಯೄ ಯೆ ಯೇ ಯೈ ಯೊ ಯೋ ಯೌ ಯಂ ಯಃ<br>
ರ ರಾ ರಿ ರೀ ರು ರೂ ರೃ ರೄ ರೆ ರೇ ರೈ ರೊ ರೋ ರೌ ರಂ ರಃ<br>
ಲ ಲಾ ಲಿ ಲೀ ಲು ಲೂ ಲೃ ಲೄ ಲೆ ಲೇ ಲೈ ಲೊ ಲೋ ಲೌ ಲಂ ಲಃ<br>
ಳ ಳಾ ಳಿ ಳೀ ಳು ಳೂ ಳೃ ಳೄ ಳೆ ಳೇ ಳೈ ಳೊ ಳೋ ಳೌ ಳಂ ಳಃ<br>
ವ ವಾ ವಿ ವೀ ವು ವೂ ವೃ ವೄ ವೆ ವೇ ವೈ ವೊ ವೋ ವೌ ವಂ ವಃ<br>
ಶ ಶಾ ಶಿ ಶೀ ಶು ಶೂ ಶೃ ಶೄ ಶೆ ಶೇ ಶೈ ಶೊ ಶೋ ಶೌ ಶಂ ಶಃ<br>
ಷ ಷಾ ಷಿ ಷೀ ಷು ಷೂ ಷೃ ಷೄ ಷೆ ಷೇ ಷೈ ಷೊ ಷೋ ಷೌ ಷಂ ಷಃ<br>
ಸ ಸಾ ಸಿ ಸೀ ಸು ಸೂ ಸೃ ಸೄ ಸೆ ಸೇ ಸೈ ಸೊ ಸೋ ಸೌ ಸಂ ಸಃ<br>
ಹ ಹಾ ಹಿ ಹೀ ಹು ಹೂ ಹೃ ಹೄ ಹೆ ಹೇ ಹೈ ಹೊ ಹೋ ಹೌ ಹಂ ಹಃ<br>
ಕ್ಕ ಖ್ಖ ಗ್ಗ ಘ್ಘ ಙ್ಙ ಚ್ಚ ಛ್ಛ ಜ್ಜ ಝ್ಝ ಞ್ಞ ಟ್ಟ ಠ್ಠ ಡ್ಡ ಢ್ಢ ಣ್ಣ ತ್ತ ಥ್ಥ ದ್ದ ಧ್ಧ ನ್ನ ಪ್ಪ ಫ್ಫ ಬ್ಬ ಭ್ಭ ಮ್ಮ<br>
ಯ್ಯ ರ್ರ ಲ್ಲ ಳ್ಳ ವ್ವ ಶ್ಶ ಷ್ಷ ಸ್ಸ ಹ್ಹ ಕ್ರ ಖ್ರ ಗ್ರ ಘ್ರ ಙ್ರ ಚ್ರ ಛ್ರ ಜ್ರ ಝ್ರ ಞ್ರ ಠ್ರ ಡ್ರ ಢ್ರ ಣ್ರ<br>
ತ್ರ ಥ್ರ ದ್ರ ಧ್ರ ನ್ರ ಪ್ರ ಫ್ರ ಬ್ರ ಭ್ರ ಮ್ರ ಯ್ರ ರ್ರ ಲ್ರ ಳ್ರ ವ್ರ ಶ್ರ ಷ್ರ ಸ್ರ ಹ್ರ ೦ ೧ ೨ ೩ ೪ ೫ ೬ ೭ ೮ ೯<br>
ಶ್ಟು ಷ್ಟು ಕ್ಟ್ ಗ್ಧ ದ್ಧ ದ್ಧೆ ಸ್ದೇ ದ್ಭು ಶ್ರೀ ಸ್ತ್ರೀ ಪ್ರ ಸ್ರ ಣ್ಯ ಧ್ಯ ವ್ಯಾ ಪ್ಯಾ ಲ್ಯ ತ್ಯ ತ್ತ್ಯನ್ಯೋ ಸ್ಯಾನ್ಯಃ ರ್ಣಾ ರ್ತೃ ರ್ಬ ತ್ತ್ವ ಶ್ವೇ ಶ್ವ ಸ್ವ ದ್ವಾ ದ್ವಾ ಸ್ವಾ ದ್ವ ಜ್ಞಾ ಜ್ಞ ಬ್ರ ಹ್ಮಾಂ ನ್ಮ ಹ್ಮ ಲ್ಮೀ ಸ್ಪಶ್ನ ಸ್ತ ಯ್ತು ಪ್ತ ಷ್ಟ್ರೀ ಸ್ಟೆ ಫ್ಲಾ ಕ್ಟ್ರಿ ಷ್ಠೆ ಶ್ಚಾ ತ್ಯ ಕ್ಪ್ಮ ಜ್ಞ ಷ್ಟ್ರ ಕ್ಷ್ಮಿ ಸ್ಥಿ ಷ್ಟ್ರಾ ವೈ ದ್ಯ್ಕ ನ್ಕೈ<br>
ಕ್ಷಂ ಕ್ಷ ಕ್ಲಾ ಥ್ಯಾಂ ಕ್ಸ್. ಹ್ಯಾಂ ಲ್ಮೇ ಲ್ಡ<br>
<br>
<br>
<br>
<br>
<br>
<br>----------<br><span><font color="#000000">From: <b>74yrs old</b> <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>><br>Date: Thu, May 8, 2008 at 3:40 PM<br>
To: Basil Brassily Thoppil <<a href="mailto:basilbthoppil@gmail.com" target="_blank">basilbthoppil@gmail.com</a>><br></font><br><br></span><div>Basil,<br>I like to have your feedback of your experience <br>
Best Wishes.<br><br><div class="gmail_quote">
2008/5/3 74yrs old <<a href="mailto:withblessings@gmail.com" target="_blank">withblessings@gmail.com</a>>:<div><div></div></div></div><br>
</div></div><br>
</div><br>
</div><br><br>
--~--~---------~--~----~------------~-------~--~----~<br>
സ്വതന്ത്ര മലയാളം കമ്പ്യൂട്ടിങ്ങ് - എന്റെ കമ്പ്യൂട്ടറിന് എന്റെ ഭാഷ
<br> പിരിഞ്ഞു പോകാന്: smc-discuss-unsubscribe@googlegroups.com
<br> സംരംഭം: https://savannah.nongnu.org/projects/smc <br> -~----------~----~----~----~------~----~------~--~---<br>
<br>