Nowadays, there are huge amount of texts written in Amharic language. Theses texts, documents and related Amharic literatures are usable for individuals, who can read, hear and understand Only Amharic languages. But In Ethiopia, there are many individuals who cannot hear and understand any text and literature written in Amharic language unless there is parallel translation in language they are good. These documents need to be translated to Awngi to provide valuable information for Awngi language speakers. To conduct the research, the corpus was collected from Amharic texts, Mass Media Agency and Bible. We used minimum of 1500 simple sentences, 1000 compound and 1000 complex sentences and maximum of 5000 sentences for each sentences type in order to train the system. We used 9:1 ratio for training and testing respectively. For language model we used minimum of 5700 and maximum of 14491 monolingual sentence of Awngi language. To do the system, we used Moses for Mere Mortal for translation process, MGIZA++ for alignment and IRSTLM for language model. Experimental results showed that better performance of 37% BLUE score was registered using complex sentences. In Amharic language, a word in sentences can have more than one meaning. While translating, the challenge of this study was not translating the meaning of the given sentences according to the context. But this study has not solved that challenge which needs further study to show all meanings of word depending on the context properly

Awngi, Awngi translation, statistical, Amharic traslation,amharic-awngi translation

[1]. Raymond J. Mooney,”CS 343: Artificial Intelligence Natural Language Processing”, University of Texas , Texas, pp 1-4 ,2018.
[2]. Abhimanyu Chopra, Abhinav Prashar, Chandresh Sain. International journal of technology enhancements and emerging engineering research, vol 1, issue 4.pp.131 – 133, 2013
[3]. Karen Louise Smith. “The Translation of Advertising Texts: A Study of English Language Printed Advertisements and their Translations in Russian.” PhD. thesis. 2002.
[4]. Esubalew Asmare Desta. “Developing Awngi-Amharic cross Language information retrieval (Clir):A Dictionary Based Query Translation Approach.”M.Sc.thesis, University of Gondar, Ethiopia, 2015.
[5]. Sisay Adugna Chala. “English – Afaan Oromoo Machine Translation: An Experiment Using Statistical Approach.” M.Sc.thesis, Addis Ababa University, Ethiopia, 2009.
[6]. Eleni Teshome.” Bidirectional English-Amharic Machine Translation: An Experiment using Constrained Corpus.”M.Sc.thesis, Addis Ababa University, Ethipia,2013.. Yitayew Solomon.”Optimal Alignment for Bi directional Afaan Oromo-English Statistical MachineTranslation”. Msc.Thesis, Addis Ababa University, Ethiopia,
[7]. June, 2017
[8]. Tsegeye Misikir,”Developing steeming algorithm for Awngi text: A longest much approach”.Msc.Thesis, Addis Ababa University, Ethiopia, 2003
[9]. John Hutchins: Reflections on the history and present state of machine translation, University of East Anglia.
[10]. Llu´ıs M`arquez Villodre, “Empirical Machine Translation and its Evaluation”, PhD.Thesis. Universitat Polit`ecnica de Catalunya, Barcelona, Maig de 2008.
[11]. John Hutchins,A new era in machine translation research: University of East Anglia, Norwich, England, pp.211-219], 1995.
[12]. Abiola O.B, Adetunmbi A.O, Oguntimilehin. A. A review of the various approaches for text to text machine translations, Afe Babalola University, Ado-Ekiti, Nigeria.
[13]. John Sturdy DeNero. “Phrase Alignment Mo dels for Statistical Machine Translation.” PhD.thesis, University of California, Berkeley, 2010.
[14]. Jabesa.” Bidirectional English-Afaan Oromo Machine Translation”. Msc. Thesis. Addis Ababa University, Ethiopia, 2013.
[15]. Nakul Sharma, English to Hindi Statistical Machine Translation System, MSC thesis, Thapar University Patiala, June 2011.
[16]. Mulu, Besacier.” English-Amharic Statistical Machine Translation”. PhD. Thesis. Addisababa University, Ethiopia.
[17]. Ruchika Sinhal and Kapil Gupta, Language Processing for MT: Need, Problems and Approaches, International Journal of Engineering Research and General Science Volume 3, Issue 5, 2015.
[18]. Chris Callison-Burch , “Machine translation:benefits and advantages of statistical machine translation and NRC’s Portage”, pp. 1-14, 2005.
[19]. Melese Mihret.” Sentiment Analysis Model for Opinionated Awngi Text”.Msc. Thesis, University of Gondar. Ethiopia, 2017.
[20]. K. Dwivedi and P. P. Sukadeve, “Machine Translation System Indian Perspectives”, Proceeding of Journal of Computer Science Vol. 6 No. 10. pp 1082-1087, May 2010.
[21]. M. D. Okpor. Machine Translation Approaches: Issues and Challenges. Journal, Vol. 11, Issue 5, No 2, September, 2014.
[22]. Khan Md. Anwarus Salam. , “Independent Study Report: Improving Example Based English to Bengali Machine Translation using WordNet”, January 2009.
[23]. Mulu Gebreegziabher, Laurent Besacier. “English – Amharic Machine Translation: An Experiment Using Statistical Approach.” PhD. Thesis, Addis Ababa University, Ethiopia.
[24]. John Sturdy DeNero. “Phrase Alignment Mo dels for Statistical Machine Translation”, PhD.thesis, University of California, Berkeley,2010.