Machine Translation In English To Hindi

1559 Words7 Pages
Abstract— India is a large polyglot country; different states have the different regional languages. English is a universal and Hindi is the National Language of Indians. It is not mandatory that each one understands English language and the speakers expect to convey their ideas and messages to their listeners .Machine translation is an area to increase the knowledgeable society of Indians without any language barrier. Machine Translation is a sub division of Natural Language Processing that focuses on translating text from one language to another. We propose a system design which uses English as the source language and Hindi language as the target. We describe and compare the approaches of Statistical and Rule based machine translation over…show more content…
MANTRA MT-system
Machine assisted translation tool (MANTRA) translates a document from English to Hindi in a specified domain of personal administration, office order, and circulars. Mantra uses Tree Adjoining Grammar for generation and parsing. It takes text as input or output of speech recognition program.
B. MATRA MT-system
MATRA, human-aided transfer based system for English to Hindi translation. The system has been applied in the domains of news, technical phrase.
C. UNL Based English to Hindi MT system
The Indian Institute of Technology has developed a machine translation system for translating from English to Hindi using the Universal Networking Language (UNL) that is interlingua. The UNL is an international project which aims to create an Interlingua for all major human languages. IV SYSTEM OVERVIEW
Here we propose a HMT Architecture for English to Hindi translation. English to Hindi machine translation can be done with the help of a bilingual dictionary of English to Hindi and a set of grammar rules. The architecture components are explained ahead in Section V. V METHODOLOGY
Here we are using a hybrid technology, encompassing the declension and rule based machine translation
…show more content…
Parse: Parsing the set of words
Parsing is a process which is used to understand the syntax and semantics of the source language. Parsing is done by using Stanford parser. The tagged input sentences is passed through Stanford parser. And obtain the grammatical rule of each sentences. C.POS taggers
Parts of speech tagging for the parsed input source sentence is done using Stanford Pos tagger. The POS tagger can identify the information about the input sentences’ words like verb, noun, adjective etc. Tagging is done by loading a “trained” file that contains the information for the tagger to tag the input words. The tagged words are passed on to the declension tagger.Below listed are some of the abbreviations of part of speech in English.D.Declension Tagger
This module takes the tagged output from the POS tagger. It will re-tag the tagged words based on the following declension rules::Re-ordering
Indian languages are the free word-order language[1].The major difference between English and Indian languages are English follows Subject-Verb-Object (SVO) format. But other Indian languages are follows Subject-Object-Verb(SOV).So reordering is necessary while different language translate to Indian languages by using predefined

More about Machine Translation In English To Hindi

Open Document