News and press releases - OCR and digitization software for Hindi and Sanskrit

Home > News

News and additional content

January 5th, 2016

Version 1.0.0.1 of GujaratiOCR released

We have added a new Indian language - the first version of GujaratiOCR has just been released.

August 4th, 2015

Version 1.0.0.4 of MarathiOCR released

ind.senz has released a new version of MarathiOCR. This release has higher recognition accuracy, a new layout analyzer, and offers improvements in the user interface.

April 30th, 2015

Version 1.0.0.2 of TamilOCR released

ind.senz has released a new version of TamilOCR. New features include higher recognition accuracy, a new layout analyzer, and improvements in the user interface. In addition, all improvements developed for the last releases of HindiOCR and SanskritOCR have been integrated in this build.

March 23rd, 2015

Version 1.0.0.1 of the Hindi OCR SDK released

ind.senz has released the first version of the software development kit (SDK) for Hindi OCR. With this tool, you can integrate OCR support for Hindi into your own C++ and C# applications.
Read more about our new Hindi OCR SDK.

January 21st, 2015

Final decisions: Free licenses of SanskritOCR for the preservation of Sanskrit literature

The call for applications was quite a success - we received over 50 applications from all over the world. After intensive consideration, the sponsor of this call finally decided to award over 30 licences to organizations and inviduals working in the field of Sanskrit digization.

For building up a network of volunteers working on the digitization of Sanskrit, the following finalists agreed to have their names and mail addresses published:

Dr. V.R. Anil Kumar (veppatangudi@gmail.com); A. Samantaray (arindra@sriaurobindoashram.org.in); B. Krishna Bal (bal@ku.edu.np; http://ltk.org.np/); B. Krishnamurthy (balaji1980@gmail.com); Prof. B. Dev (brahma63@gmail.com); D. Patel (drdhaval2785@gmail.com); Dr. D. Mani (dindayalmani@gmail.com); Dr. D.K. Rana (director@chinfo.org); Prof. G.N. Jha (girishjha@gmail.com); G.P. Shastri (gyanprakashshastri@gmail.com); Prof. J.D. Sathe (jayashree.sathe@gmail.com); M. Gluckmann (martin@vedicsociety.org); P. Hackett (ph2046@columbia.edu); P. Chaitanya (pranipatachaitanya@yahoo.co.in); Dr. P. Ramanujan (ramanujachar@gmail.com); Prof. R. Bharadwaj (rameshbharadwaj85@yahoo.com); R. Srirangam (r_srirangam@hotmail.com); S. Mishra (sampadanandamishra@gmail.com); Dr. S. Nanal (srnanal@yahoo.com); S. Brahma (shrivathsa.brahma@gmail.com); S. Thyagarajan (shivakumar24@gmail.com); Sringeri Publications (publications@sringeri.net); S. Thompson (revdocstephen@gmail.com); S. Chandra (subhash.jnu@gmail.com); Venkatasubramanian (venkat.s@vyomalabs.in); V. Agarwal (virendra.agarwal@vedicgranth.org); V. Bhide (vrbhide@hotmail.com); A. Pugliese (angelo.pugliese@yahoo.com); Y. Mehta (yashwant.mehta@gmail.com); Vladimir (charu67@mail.ru)

December 4th, 2014

Call for application: Free licenses of SanskritOCR for the preservation of Sanskrit literature

We are very pleased to announce that due to the funding proposal of a generous sponsor, we are able to offer 15 free licenses of SanskritOCR to those who are residing in India and 5 free licenses of SanskritOCR to those who are residing in other countries. The licenses will be awarded till Vasant Panchami to those who are dedicated towards the preservation of Sanskrit scriptures and literature.

The software licenses will be awarded based on the following information. Please note that we can only accept applications if the complete information is provided.

Please provide the information below.

Your name, e-mail address, and mail address with country (Telephone # optional)
Name of affiliated organization or website, if any
Type of projects you are working in for Sanskrit Language preservation
Have you used this or similar software before?
Have you performed any Sanskrit preservation work before? If yes, can you describe it?
On which particular scriptures or books are you planning to work for conversion and preservation? Any thematic focus of your preservation work?
Are you planning to make the digitized Sanskrit texts publicly available, e.g., by uploading the digital texts into a web repository?

The deadline for application is December 25, 2014. Applications should be sent by this date at latest to oliver.hellwig@indsenz.com

The awarded parties will be notified by mail when the sponsor has made the final decision about the applications. The software licenses and all accompanying information will be provided as downloadable files.

ind.senz will honor software related support at the same level as available with the regular purchase. This means that licenses are time-unlimited, single-place licenses without digitization quota and with two years of free upgrades.

In case you know someone who may be interested in this call, please do not hesitate to forward this mail.

Disclaimer: The sponsor’s decision is final and is not subject to review or appeal.

November 19th, 2014

Version 1.0.1.0 of SanskritOCR released

ind.senz has released a new version of the OCR program for Sanskrit. New features of this release include:

The recognition quality and the lexicographic support for Sanskrit have been improved.
The OCR engine handles words with Sandhi in a more consistent and reliable manner than in previous versions.
The scanning support has been improved.
SanskritOCR integrates the new spellchecking procedure introduced with the previous version of the Hindi OCR.
Copy/paste are activated in the recognized text.

We are constantly tuning and improving the OCR engine. As a showcase for the quality of the OCR, we built a database of searchable scanned Sanskrit books that you find at www.sanskrit-linguistics.org.

September 12th, 2014

Version 1.0.1.1 of HindiOCR released

ind.senz has released a new version of HindiOCR. Features of this release include:

Higher recognition accuracy
We have redesigned and extended the spellchecker. In case an unclear word is found, the spellchecker proposes similar words from the dictionary.
Copy/paste activated in the output text window.
Transliteration of Hindi according to the ISO 15919 standard

May 28th, 2014

Version 1.0.0.3 of MarathiOCR released

ind.senz has released the third version of MarathiOCR. New features in this version include, among others:

Better dictionary and language support for Marathi
Unicode support: The OCR program accepts file names in Marathi.
Tools for faster correction of recognized text (alternative proposals, copy/paste in the output window)

April 7th, 2014

First release of TamilOCR

We have released the first version of TamilOCR engine.

December 1st, 2013

New version of SanskritOCR released

ind.senz has released a new version of the SanskritOCR engine that integrates all features of the professional version of our latest Hindi OCR engine, plus increased recognition accuracy for Sanskrit texts:

Batch processing of multiple pages
OCRing complete directories
Output of recognition results as text-under-image pdf files

October 25th, 2013

New version of HindiOCR released

ind.senz has released a new version of the HindiOCR engine. In addition to the features integrated in the previous release of MarathiOCR, the professional version of HindiOCR now produces text-under-image PDF output (searchable PDFs), and comes with a special function for processing complete directories in a "silent", unsupervised mode. Existing batch files can be completely removed from disk with a newly added feature.

October 1st, 2013

Version 1.0.0.2 of MarathiOCR released

ind.senz has released an upgrade of the OCR engine for Marathi. New features include:

Improved Marathi dictionary and classifiers
Better layout analysis
Built-in support for ISCII typing and for the standard English keyboard

In addition, we fixed some stability issues that were reported for the previous release.

June 19th, 2013

Version 1.0.0.9 of HindiOCR released

ind.senz has released a major upgrade of its OCR engine for Hindi, featuring a number of improvements:

Higher recognition speed
Better support for Nagari numbers
Customizable user dictionary with import function
"Auto increase" feature for small text
... and the new user interface introduced with the first release of MarathiOCR.

May 5th, 2013

Version 1.0.0.1 of ind.senz MarathiOCR engine released

ind.senz has released the first version of Marathi OCR, continuing the development of OCR engines for Devanagari script.

MarathiOCR digitizes printed Marathi documents with high recognition accuracy and speed. The program contains built-in classifiers for about 200 Marathi-Devanagari letters and lexicographic information for modern Marathi. Recognized text is displayed in Unicode-Devanagari encoding, and can be exported into a number of file formats including Unicode-RTF, plain Unicode text, and various transliteration schemes.

As in HindiOCR and SanskritOCR, MarathiOCR provides a training option for unusual and rare fonts. The program processes standard image formats (bmp, jpg, png, tiff, gif) and works on Windows XP® and Windows 7®.

MarathiOCR is built with the same core as HindiOCR version 1.0.0.8. The program comes with a new, customizable user interface, and outperforms previous versions in terms of speed and recognition accuracy.

February 27th, 2013

Version 1.0.0.8 of ind.senz HindiOCR engine released

ind.senz has released a new version of its OCR engine for Hindi.
Improvements and new features:

Improved layout analysis
Higher recognition accuracy due to better lexical support and new classifiers
Online feedback about the expected recognition quality while processing a page
Dictionary interface in Devanagari
New methods for increasing the recognition quality on degraded text pages

December 21st, 2012

Version 1.0.0.8 of ind.senz Sanskrit OCR engine and iTunes GitaReader released

ind.senz has released a new version of its OCR engine for Sanskrit. Improvements and new features:

Improved layout analysis
Higher recognition accuracy due to better lexical support and new classifiers
Smaller memory footprint
Online feedback about the expected recognition quality while processing a page
Dictionary interface in Devanagari
New methods for increasing the recognition quality on degraded text pages

GitaReader for iTunes

Sven Sellmer has developed an iPhone app that gives you a new, in-depth view of the Bhagavadgita. Get the GitaReader through iTunes, or read more about the features of this app.

October 5th, 2012

Version 1.0.0.7 of ind.senz OCR engines and a Sanskrit smartphone app released

ind.senz has released new versions of the OCR engines for Hindi and Sanskrit. The OCR programs provide extended data input and data export features:

Recognition:

New recognition mode: More accurate recognition for trained fonts.
Higher recognition accuracy for Latin punctuation marks and brackets

Keyboard and input:

Built-in keyboard layout allows input of all Devanagari Unicode letters.
User defined keyboard layouts
ENTER key working in recognized text window

Text export:

Justified text blocks can be preserved during export.
(Professional version:) Store multiple pages in separate files
Export recognition results as a pure text file without RTF formatting and in Kruti Dev formatting

First "Sanskrit at your fingertip" app released

ind.senz presents a new generation of smartphone applications for Indian languages: Sanskrit at your fingertip provides classical Sanskrit texts with rich grammatical and lexicographic annotation and Sandhi resolved, in an interactive and smart design.

Read more about the background of the series and about GitaReader, which features the full text of the Bhagavadgita.

August 9th, 2012

Version 1.0.0.6 of ind.senz OCR engines released

ind.senz is proud to announce the release of the new, more powerful OCR engines for Hindi and Sanskrit. The OCR programs present themselves in a new, more user friendly design. They come with a bunch of new functions, incorporating valuable feedback of our users.

HindiOCR is now available in a basic version and a professional version for more demanding digitization tasks. Please have a look at the product page for further information.

New features:

Optimized user interface with a more comfortable navigation
Support for TIFF and GIF images
Improved layout analysis
Recognized text can be displayed in any Devanagari-enabled Unicode font installed on your computer.
Improved memory management
Natural sorting order for imported images, as found in Windows Explorer
Recognition speed increased by up to 15% when compared with former versions
Better lexicographic coverage of recognized Hindi through a larger dictionary

Preview: Later this year, we will start publishing a series of Sanskrit apps for iPhone and Android, making available the most popular works of Sanskrit literature with rich additional content on mobile devices. Just subscribe to our newsletter to stay informed!

May 8th, 2012

Tutorial: How to optimize recognition results

Link

April 19th, 2012

Version 1.0.0.5 of OCR engines released

ind.senz has released versions 1.0.0.5 of HindiOCR and SanskritOCR. The new versions contain numerous improvements and new features:

Automatic layout analysis for standard rectangular layouts
New set of classifiers with better recognition accuracy
Higher speed and optimized memory usage
Text export in Latin transcription, optionally in IAST, Harvard-Kyoto, ITrans and Velthuis transliteration
HindiOCR: Better support for Latin numbers in Devanagari text
SanskritOCR: Improved linguistic postprocessing with reduced error rates

Please check out the demo versions for the new features!

April 12th, 2012