NEWS and EVENTS

• CRBLP research wins BASIS IT Innovation Search Award in SOFTEXPO 2010
  Katha – a CRBLP-developed software to translate Bangla text to speech was chosen as the Best Innovative Project during BASIS SOFTEXPO 2010. Bangladesh Association of Software and Information Services (BASIS), the national trade Association of software and IT enabled services companies of Bangladesh, recently held its annual exposition BASIS SOFTEXPO 2010 at the Bangabandhu International Conference Centre. One of the major events at the expo this year was an IT Innovation Search Program designed to discover and reward local innovations in IT. The BASIS jury had selected 3 finalists, two of which were BRACU CRBLP research outputs, to highlight during the expo. The other finalist from CRBLP was Bangla OCR (Optical Character Recognition) software that can convert images of printed Bangla text to editable text. The Katha research team consists of Firoj Alam, S. M. Murtoza Habib, Kamrul Hayder and Naira Khan; the Bangla OCR team consists of Md. Abul Hasnat, Muttakinur Rahman Chowdhury, S. M. Murtoza Habib and Dil Afroza Sultana. Professor Mumit Khan, Head of the Centre for Research on Bangla Language Processing (CRBLP), received the award on behalf the CRBLP research team at the ceremony.
  
  

• Microsoft Bangladesh organized a certificate distribution ceremony on 18th January 2010 at Spectra convention center, Gulshan, Dhaka. The country manager Firoj Mahmud distributed certificates among the CRBLP team members of Microsoft Windows Vista and Office 2007 Bangla localization projects.
  
  
  
  
• S. M. Murtoza Habib and Firoj Alam attended International Workshop on Spoken Language Prosody (IWSLPR-09) organized by Centre for Development of Advanced Computing (C-DAC), Kolkata, India, November 25 – 27, 2009.
  
• S. M. Murtoza Habib attended and presented on "Developing Bangla text, speech and script resources: current status and challenges ahead" in Asian Language Resource Summit (http://www.tcllab.org/alrs/), organized by Thai Computational Linguistics Laboratory (TCLLAB) and National Institute of Information and Communications Technology (Japan), held at Evason Phuket & Six Senses Spa, Phuket, Thailand from March 20 - 21, 2009.
  
  

• The Center for Research on Bangla Language Processing (CRBLP) at BRAC University launched the first official release of the Bangla language processing software packages for Text to Speech (TTS) and Optical Character Recognition (OCR) named “Katha” and BanglaOCR respectively, on February 19, 2009 at 3 pm at BRAC University.
  The chief guest at the launching ceremony, Dr. Salehuddin Ahmed, Pro Vice Chancellor, BRAC University spoke at the occasion, along with Monsur Ahmed Choudhury, MD, Jatyo Protibondhi Unnayan Foundation (JPUF), Saiful Islam Khan, Project Officer, Joint project ‘Strengthening Government Integrated Education program for the Blind’ of the Shomaj Sheba Odibhoptor and Sight Savers International, and Dr. Mumit Khan, Professor, CSE, Head, CRBLP, BRAC University. Guests included Shamsur Nahar Ahmed, Sight Savers International and Rafiq Zaman, National Forum of the Organizations Working with the Disabled.
  
  CRBLP has been working on generating speech from Bangla Text. Text To Speech (TTS) system for the Bangla language can help tackle the illiteracy problem, empower the visually impaired and increase the possibilities of improved human-machine interaction. Optical Character Recognition (OCR) is the process of converting printed text images to editable text. CRBLP has been developing a Bangla OCR that takes the scanned image of a printed page or document as input and converts it into editable Unicode text. CRBLP is working on better integration with screen readers in collaboration with the vision impaired community. Although these software exist for other languages, this was the first release for Bangla in Bangladesh and represents a substantantial leap in including the visually impaired in bridging the digital divide.
  
  The TTS and OCR run on Linux, Windows and Mac OSX. There is also a web-enabled front-end for the TTS (and under development for the OCR), making these tools available at anytime and from anywhere.
  
  
  Web version of Katha: Bangla Text to Speech demo is available in online: http://www.bracu.ac.bd/research/crblp/demo/tts
  
  
• Md. Abul Hasnat, Research Programmer at CRBLP, has been awarded the Erasmus Mundus Scholarship for the European Masters Program in LCT “Language and Communication Technology” (EM-LCT) and Graduate Master Course CIMET “Colors in Informatics and Media Technology”.
  
  
• Firoj Alam, Farhana Faruqe and Muttakinur Rahman Chowdhury (Shouro) participated and presented "Language resource technologies available in Bangla" in Asian Applied Natural Language Processing for Linguistics Diversity and Language Resource Development (ADD4), NECTEC, Bangkok, Thailand, February 23-27, 2009.
  
  
• Md. Abul Hasnat and Firoj Alam attended on the Conference of Language and Technology (CLT09), held at NUCES, Lahore, Pakistan. January 22-24, 2009.
  
  
• Professor Mumit Khan and Md. Abul Hasnat attended a meeting of PAN Localization Project in Laos, January 12-16, 2009.
  
  
• CRBLP conducted a seminar on “Bayesian Approach” by Kazi Saidul Hasan. The seminar held at January 5, 2009 from 11:30 am – 12:30 pm in AH 803, BRAC University. Kazi Saidul Hasan is a Ph.D. Student, Department of Computer Science, University of Texas at Dallas. [presentation]
  
  
• CRBLP has released a new alpha version of the BanglaOCR based on Tesseract's recognition engine for both Windows and Linux under the GNU Public License (GPL) version 2. [download]
  
  
• A paper accepted in 6th International Conference on Natural Language Processing (ICON-2008) conference on "Detecting flames and insults in text" by Altaf Mahmud, Kazi Zubair Ahmed and Mumit Khan, CDAC Pune, India, December 20 - 22, 2008.
  
  
• A paper accepted in International Joint Conferences on Computer, Information, and Systems Sciences, and Engineering (CISSE 08) conference on "BWN - A Software Platform for Developing Bengali WordNet" by Farhana Faruqe and Mumit Khan.


• A paper accepted in CLT09 conference on "Text Normalization system for Bangla" by Firoj Alam, S. M. Murtoza Habib and Mumit Khan.


• A paper accepted in CLT'09 conference on "Integrating Bangla script recognition support in Tesseract OCR" by Md. Abul Hasnat, Muttakinur Rahman Chowdhury and Mumit Khan.


• A paper accepted in DRR XVI conference on "Rule based segmentation of lower modifiers in complex Bangla scripts" by Md. Abul Hasnat and Mumit Khan.


• A paper accepted in DRR XVI conference on "Elimination of splitting errors in printed Bangla scripts" by Md. Abul Hasnat and Mumit Khan


• Unicode released CLDR 1.6.0. For Bengali (Bangladesh) part of CLDR data, the contributors include S. M. Murtoza Habib, Firoj Alam, Naira Khan, Kamrul Hayder and Mumit Khan from CRBLP. The CRBLP team submitted data with the concern of IBM India and US.


• CRBLP team members (MD. Abul Hasnat, S.M. Murtoza Habib, Farhana Faruqe, Firoj Alam) attended in the Summer School on Natural Language Processing and Text Mining workshop held at IIT Kharagpur from 24 to 29 June, 2008. They also visited Media Lab Asia Research Laboratory at IIT Kharagpur and Computer Vision and Pattern Recognition (CVPR) unit at Indian Statistical Institute (ISI).


• A paper accepted in Interspeech 2008 conference on "Text Normalization system for Bangla" by Firoj Alam, S. M. Murtoza Habib and Mumit Khan. [Withdrawn]


• The Global Information Society Watch (2007) report covered activities of CRBLP saying, "BRAC University now hosts the country's main localisation centre". ( read the original report: pages 114-115). [download]


• CRBLPConverter V1.1 Released: CRBLP released CRBLPConverter version 1.0, under the GNU Public License 2 with some new features and bug fixes. Source and binary for CRBLPConverter can be downloaded directly from sourceforge. It is platform independent, so it can run to any OS. Development team of the CRBLPConverter includes Fahim Tawfique Chowdhury, Md. Zahurul Islam, and Mumit Khan. download
  
  
• Naushad UzZaman, Research Programmer, has received Computer Science PhD offers from University of Rochester (NY, USA), University of Pittsburgh (PA, USA) and University of Texas at Dallas (TX, USA); Computer Science MS offers from Columbia University (NY, USA), University of Alberta (Canada) and offer for Masters program on Computer Science and Language Technology, funded by European Union with Erasmus Mundus scholarship. Naushad ended up going to University of Rochester for his PhD from Fall 2007.
  
  Md. Zahurul Islam, Research Programmer, has received Computer Science PhD offers from University of North Texas (TX, USA), and offer for Masters program on Computer Science and Language Technology, funded by European Union with Erasmus Mundus scholarship. Zahurul ended up going to University of Saarland, Germany for his Masters from Winter 2007.
  
  Md Nafid Haque, Research Intern for Summer 2006 at CRBLP, has been awarded the Erasmus Mundus Scholarship for the European Masters Program in Language and Communication Technology (EM-LCT 2007-2009). He will be joining  Rijksuniversiteit Groningen, The Netherlands (University of Groningen) for his first year of the 2 years masters program. For his second year, he will be joining University of Malta, Malta, another partner university under the LCT Consortium.
  
  Naira Khan left for the U.S.A. to attend the LSA Summer Institute 2007. Formed in 1924 the Linguistic Society of America (LSA) is an organization devoted to the scientific study of human language, and is the largest professional society for linguistic researchers all over the world. The LSA sponsors a biennial Linguistic Institute - a four-week summer program accompanied by a range of conferences and workshops. The classes and meetings draw researchers and graduate students from all over the world. Each Institute features leading faculty in the field, and honors the contributions of particular linguists with their appointment by the LSA to titled chairs at the Institute: the Sapir chair in general linguistics, the Collitz Chair in historical linguistics, and (from 2005) the Ken Hale chair in linguistic fieldwork and the preservation of endangered languages. The Institute of 2007 titled 'Empirical Foundations for Theories of Language' is being hosted by the Department of Linguistics at Stanford University and features 18 presession courses and 74 regular session courses. The Institute is being co-sponsored by the North American Association for Computational Linguistics and the American Dialect Society.
  
  
• CRBLP conducted a seminar on “Towards Building Emotionally Sensitive Computers” by Mohammed Ehsan Hoque. The seminar held at July 19, 2007 from 3:30 pm – 5:00 pm in UB 104, BRAC University, 66 Mohakhali, Dhaka-1212.
  "Towards Building Emotionally Sensitive Computers"
  Ever wondered whether computers will ever able to understand and respond to our emotions? Ever wondered whether it is possible for computers to understand our emotion simply from tone of our voice?
  In this seminar he talked about how we can leverage techniques from computer science, psychology and linguistics towards building machines that understand human emotions better. He will also expand his future direction of research of building computational tools to cure “Autism” at the MIT Media Lab as a PhD student.


• CRBLPConverter V1.0 Released: CRBLP released CRBLPConverter version 1.0, under the GNU Public License 2. Source and binary for CRBLPConverter can be downloaded directly from sourceforge. It is platform independent, so it can run to any OS. Development team of the CRBLPConverter includes, Md. Zahurul Islam, Firoj Alam and Mumit Khan. download
  
  
• CRBLP signed an MOU with DU
  CRBLP signed an MOU with the Linguistics Department of the University of Dhaka to build a research partnership with the goal of increasing our national capacity in the field of Computational Linguistics. A key component is to develop the underlying linguistic framework that is needed to further the Computational Linguistics work being done at CRBLP; another is to jointly develop Bangla linguistic resources such as lexica, corpora and dictionaries. To kick off the research partnership, CRBLP and DU Linguistics will jointly conduct a series of workshops, starting with topics ranging from Bangla Morphology, Phonology and Acoustic Phonetics to Corpus Analysis. Dr. Mumit Khan, Head of CRBLP and Dr. Feroza Yasmin, Chairperson of DU Linguistics, signed the MOU on behalf of their respective institutions. The BRACU team consisted of Mr. Matin Saad Abdullah and Mr. Firoj Alam, and the DU team included Dr. Jinnat Ali and Mr. Hakim Arif among others.
  
  
• During BDOI 2007, Bangladesh Olympiad in Informatics at BRAC University, Dr. Muhammad Jafar Iqbal, Dr. Muhammad Kaykobad and few other people visited CRBLP to learn about their various activities on Bangle language processing at the centre. Few of them dropped their comments seeing activities of CRBLP. [Dr. Muhammad Jafar Iqbal's comment] [Dr. Muhammad Kaykobad's comment]
  
  
• CRBLP released first "test" version of the HMM-based Bangla OCR, under the GNU Public License. Source and binary for BanglaOCR can be downloaded directly from sourceforge <http://sourceforge.net/project/showfiles.php?group_id=158301>. It currently runs only under the .NET platform, but the developers plan for a platform independent port in the near future. Development team of the BanglaOCR includes, Md. Abul Hasnat, S M Murtoza Habib and Mumit Khan.
• CRBLP researchers Naushad UzZaman and Md. Zahurul Islam have been selected to attend the ADD-2 training in Thailand with full financial support. The workshop is sponsored by Thai NECTEC's NICT Asia Research Center. Selection in this workshop was based on applicant's educational background, career, current project experiences, intention and potential to carry out and establish NLP (Natural Language Processing) research network in the region and the quality of paper submitted to the ADD workshop.
• Dr. Mumit Khan, Head, CRBLP and Associate Professor, CSE Department, attended a meeting with other Team Leaders of PAN Localization Project in Bhutan, January 2007.
• Two papers by CRBLP members, M. Hammad Ali and Nafid Haque got accepted in Education Without Borders, 2007 Student Conference, Feburary 25-27, Abu Dhabi. One of the paper titled "A Decentralised Approach to Information Retrieval for a developing country like Bangladesh" by M. Hammad Ali and Nafid Haque has been identified by the reviewers to be among the top 36 papers of the entire conference. First author of the paper, M Hammad Ali received full scholarship for attending the conference.
• Five papers by CRBLP members, Md. Zahurul Islam, Md. Abul Hasnat, Firoj Alam, S M Murtoza Habib, Altaf Mahmud and Mumit Khan, got accepted in 1st International Conference on Digital Communications and Computer Applications (DCCA2007), Irbid, Jordan, 2007.
• James Fruchterman, President and CEO of Benetech, visited CRBLP on December 6, 2006 during his brief stopover in Bangladesh. Mr. Fruchterman has recently been awarded a MacArthur Fellowship, popularly known as the "genius grant" for his work on social entrepreneurship. He has recently been profiled by the IEEE Spectrum magazine. He met with Dr. Mumit Khan to learn of the various activities at the centre, and to discuss potential future collaborations with BRAC University.
• A paper by Munshi Asadullah, Md. Zahurul Islam and Mumit Khan was accepted by the International Conference on Natural Language Processing (ICON 2007), January 4-6 in Hyderabad, India. ICON is the premiere Natural Language Processing conference in India, and this is the first paper from Bangladesh to be accepted by ICON. Munshi Asadullah and Md. Zahurul Islam participated and presented their paper in the conference.
• Two papers by CRBLP members Fahim Muhammad Hasan, Nafid Haque, Hammad Ali, Naushad UzZaman, Matin Saad Abdullah and Mumit Khan were accepted by the International Conference on Systems, Computing Sciences and Software Engineering (SCSS 06) of International Joint Conferences on Computer, Information, and Systems Sciences, and Engineering (CISSE 06). Fahim Muhammad Hasan and Nafid Haque presented their respective papers online from the CRBLP offices on December 5, 2006.
• The 9th International Conference on Computer and Information Technology (ICCIT 2006), to be held during December 21-23 of 2006 in Dhaka, has accepted 11 papers authored by the following CRBLP team members: Munirul Mansur, Yeasir Arafat, Ayesha Binte Mosaddeque, Md. Tarek Habib, Md. Jahangir Alam, Rajib Rahman, S. M. Murtoza Habib, Nawsher Ahmed Noor, Md. Abul Hasnat, Md. Mahmudul Haque, A. M. Shoeb Shatil, Tofazzal Rownok, Md. Zahurul Islam, Naushad UzZaman, Naira Khan and Mumit Khan.
• CRBLP has received a Local Language Programme (LLP) grant from the Microsoft Corporation to localize the next version of Microsoft Windows (Vista) and Microsoft Office (Office 2007) in Bangla. The Microsoft LLP also includes Bangladesh Computer Council, who will create the glossary of computer terminology and the translation style guide, which CRBLP will use to create the Bangla Language Interface Pack (LIP). Windows Vista and Office users will then be able to download the Bangla LIP for free from Microsoft.
• Naira Khan left on a Fulbright scholarship on August 17, 2006, to teach Bengali for one academic year at Duke University, in the Critical Languages Program under the NC Center for South Asia Studies. She is also teaching at The University of North Carolina – Chapel Hill and North Carolina State University. She will be rejoining CRBLP in May, 2007.
• CRBLP offered the Summer Internship 2006 program to the students of BRAC University. 13 students have been selected to participate in the program. CRBLP staff members supervised the interns on different projects during this Internship.
• S M Murtoza Habib, Md. Abul Hasnat and Firoj Alam participated a 10 week PAN Localization training program at Pakistan, May 2006.
• Dewan Shahriar Hossain Pavel, Asif Iqbal Sarkar, Arnab Zaheen, Naushad UzZaman and Mumit Khan got their papers accepted in the International Conference on Computer Processing on Bangla (ICCPB-2006), Dhaka, Bangladesh, 17 February, 2006.
• CRBLP staff members offered Natural Language Processing course to the undergraduate students of BRAC University. This course was instructed by Mumit Khan and Naushad UzZaman (Laboratory module), CRBLP staff members supervised the semester projects of the NLP course, January 2006.
• Naushad UzZaman and Mumit Khan got their paper accepted in The Second International Conference of Mobile Technology, Applications and Systems, Guangzhou, China, November 2005.
• Naushad UzZaman participated and presented his paper (also co-authored by Mumit Khan) in the 2005 IEEE International Conference on Natural Language Processing and Knowledge Engineering, Wuhan, China, October 30 - November 1, 2005.
• Sajib Dasgupta, Dewan Shahriar Hossain Pavel, Asif Iqbal Sarkar, Naira Khan, Zahurul Islam and Mumit Khan got their papers accepted in the Proc. 8th International Conference on Computer & Information Technology (ICCIT), Islamic University of Technology (IUT), Dhaka, Bangladesh, 2005.
• Sajib Das Gupta joined the Department of Computer Science as a graduate student in the University of Texas Dallas. He will pursue his higher education on Computer Science with specialization in Natural Language Processing, August 2005.
• Naushad UzZaman participated and presented his paper (also co-authored by Mumit Khan) in The Fourth IASTED International Conference on Computational Intelligence, Calgary, Alberta, Canada, July 2005.
• Sajib Das Gupta, Asif Iqbal Sarkar and Mumit Khan participated in the training and experience sharing program organized by PAN Localization, June 2005.
• Sajib Das Gupta, Naushad UzZaman and Mumit Khan participated and presented their papers in the 7th International Conference on Computer and Information Technology, Dhaka, Bangladesh, December, 2004.

Check CRBLP picture gallery.