LoReHLT Hausa Representative Language Pack consists of Hausa monolingual text, Hausa-English parallel text, annotations, amateur web audio recordings, supplemental resources and related software tools developed by the Linguistic Data Consortium for LoReHLT, a companion project of the DARPA LORELEI program. The LORELEI (Low Resource Languages for Emergent Incidents) program was concerned with building human language technology for low resource languages in the context of emergent situations like natural disasters or disease outbreaks. Linguistic resources for LORELEI include Representative Language Packs and Incident Language Packs for over two dozen low resource languages, comprising data, annotations, basic natural language processing tools, lexicons and grammatical resources. Representative languages were selected to provide broad typological coverage, while incident languages were selected to evaluate system performance on a language whose identity was disclosed at the start of the evaluation. Hausa is spoken across western Africa; it is the official language in Niger and Nigeria.  This release is the result of a pilot effort preceding the LORELEI program. Text data was collected in the following genres: news, reference, social network, and weblogs. Both monolingual text collection and parallel text creation involved a combination of manual and automatic methods. Also collected were amateur web audio recordings related to disaster events that were covered in the text data.  Data volumes are as follows: 4.4 million words of Hausa monolingual text, over 900,000 of which were translated into English 86,000 Hausa words translated from English data 30 minutes of Hausa audio recordings Approximately 96,000 words were annotated for named entities, and over 13,000 words were annotated for full entity including nominals and pronouns. Noun-phrase chunking was applied to more than 7,400 words. Over 9,600 words were labeled with simple semantic annotation. Topic annotation was applied to the audio recordings. Lexical resources and software tools are also included in this release. The tools recreate original source data from the processed XML material, condition text data users download from Twitter, apply sentence segmentation to raw text, and support named entity tagging. Monolingual and parallel text are presented in XML with associated dtds. Annotation data is presented as tab delimited files or XML. All text is UTF-8 encoded. The audio recordings are presented in .mp4 format.

Creator(s)

Jennifer Tracey

Stephanie Strassel

David Graff

Jonathan Wright

Song Chen

Kira Griffitt

Neville Ryant

Seth Kulick

Dana Delgado

Michael Arrigo

Distributor(s)

Linguistic Data Consortium

Right Holder(s)

Portions © 2002-2007, 2009-2010 Agence France Presse, © 2000 American Broadcasting Company, © 2000 Cable News Network, LP, LLP, © 2008 Central News Agency (Taiwan), © 2010-2011, 2013-2015 China Radio International. CRI., © 1989 Dow Jones & Company, Inc., © 2015 Google LLC, © 2014 Leadership Media Group, © 2005 Los Angeles Times – Washington Post News Service, Inc., © 2012-2015 Media Trust Limited, © 2000 National Broadcasting Company, Inc., © 1995, 2005, 2006, 2010 New York Times, © 2000 Public Radio International, © 2014 RFI, © 2003, 2005-2008, 2010 The Associated Press, © 2003, 2005-2008 Xinhua News Agency, © 2013 Zongolink Radio, © 2015, 2018, 2024 Trustees of the University of Pennsylvania