Free Essay

Language Translator

In: Computers and Technology

Submitted By emahuja
Words 1593
Pages 7
International Journal of Engineering and Advanced Technology (IJEAT)
ISSN: 2249 – 8958, Volume-X, Issue-X

Language Translator
Ekta Ahuja, Karan Kashyap


degradation of translation output quality over a period of time. Abstract— The aim of this project is to find out the problem areas in the translation done by Bing, suggest a method to improve the language translation and build a translator that supports accurate English to Hindi translation. The English input can be of textual, voice or image form and the Hindi output will be in the textual format.. With our research we aim to suggest a way to improve the already available popular translation engines. For this purpose we have used the method of hybrid machine translation. Hybrid machine translation is a method of machine translation that is characterized by the use of multiple machine translation approaches within a single machine translation system. The motivation for developing hybrid machine translation systems stems from the failure of any single technique to achieve a satisfactory level of accuracy.

B. Statistical Machine Translation
Statistical Machine Translation systems make use of computer algorithms that find out many possible ways of connecting smaller pieces of text together, in order to produce a best translation. Statistical Machine translation basically translates words and phrases along with their statistical likelihood. These are learned automatically from previously translated text, creating a bilingual “database” of translations. A program referred as decoder matches the source code and phrases with the translation databases and searches for all possible translation combinations. An algorithm is then used to select the best translation out of the millions of possible translations and output it. The main advantage of statistical systems is the high levels of automation involved in developing as well as maintaining these systems. As machine-learning algorithms are used, deploying new statistical MT systems is greatly faster and more cost efficient as compared to older technologies. Also, it is easy to update this system that leads to improved performance and increased translation coverage over time.
However, these translation systems have their shortcomings as well. As automated machine learning techniques are used it becomes difficult to control the learning process and identify the source of potential errors. These systems are limited in addressing language structure when translating from source to target language.

Keywords—translator, Bing, textual, hybrid.
.

I

INTRODUCTION

Nowadays, language translation and interpreting services are useful in all spheres of daily life, like to help a student to enroll into a new school, or to interpret for a corporate executive who has just purchased a new business in India.
Translation can play a number of different roles such as a
‘unifying’ or constructing new words but also, most crucially, as a source of knowledge about foreign, lesser known cultures. Thus, translation is not merely a linguistic process, but can also make a political and social impact.
II

PROPOSED APPROACH

C. Hybrid Machine Translation
Hybrid machine translation is a method of machine translation that is characterized by the use of multiple machine translation approaches within a single machine translation system. The motivation for developing hybrid machine translation systems stems from the failure of any single technique to achieve a satisfactory level of accuracy.
In this hybrid approach, the translation is performed using a
Rule-Based MT and a Statistical MT is then used in order to adjust the output. In what seems to be a theoretically sensible approach, hybrid solutions intend to combine the advantages of the individual approaches to achieve an overall better translation. The approach is most useful to address one of Rule-Based MT greatest challenges – translation ambiguity. When a word/phrase can have more than one meaning, statistics can help identify the most suitable option. Hybrid MT systems provide quality improvement benefits over each individual approach.

A. Rule-Based Machine Translation
Rule-Based Machine Translation systems use large collections of rules that are manually developed over time by human experts. They basically map structures from the source to the target language. Rule-based MT requires manually building extensive translation lexicons. Users may often edit or manipulate them for better translations. The human skills invested in the development of rule-based systems have the potential of delivering fairly good automated translations with predictable results. It is possible to adjust the errors safely through changes to user dictionaries. However, as it involves significant manual labor, rule-based systems can be quite costly and time consuming to implement as well as to maintain. But as rules are added and updated, these systems have the potential of generating ambiguity which can cause significant


Manuscript received October 01, 2011. (Fill the Details)
Ekta Ahuja, Computer Science, Vivekanand Education Society’s
Institute of Technology, Mumbai, India, +917666877900, (email:ektaahuja19@gmail.com).
Karan Kashyap, Computer Science, Vivekanand Education Society’s
Institute of Technology, Mumbai, India, +919867185578, (email:karan.kashyap26@gmail.com).

III

REQUIREMENTS

A. Functional Requirements
The system must translate the given input text into the language specified by the user.

1

Published By:
Blue Eyes Intelligence Engineering &
Sciences Publication Pvt. Ltd.

Please Enter Title Name of Your Paper
Our module is based on the language Hindi. The translator should provide an additional feature wherein the translated text can be shared on the various social and professional platforms. It must allow three important features.
Text to Text.
Speech to Text.
Image to Text.
B. Non Functional Requirements
Loading of a user interface element in a translated format should take no longer than the acceptable standard .Currently this is set at a second to load the page. For the smooth and efficient working of the system, internet connection is must. In order to enjoy the benefits of sharing feature, device must contain the Hindi keyboard application.

.

Figure 1: System Design.

V OBSERVATION AND ANALYSIS
IV IMPLEMENTATION

A. Bing Translator
Input is: He sings song.
Statistical analysis of ‘sings ’ is done as follows

Our application is build using some available softwares and
APIs. For translation, we have made use of Bing API provided for the application development by Microsoft
Corporation. And for text recognition on the clicked image, we've made use of the tesseract software. Out of all the softwares available, tesseract is the most accurate one.
Because it stores all the possible patterns that a specific alphabet can take. So, it gives the best results as far as text recognition from a image is concerned.
Apart from these softwares, we’ve build our application with the help of android developers tools like The Android
Developer Tools (ADT) plug-in for Eclipse provides a professional-grade development environment for building
Android apps. It's a full Java IDE with advanced features to help you build, test, debug, and package your Android apps
Our Module in the app makes use of Hybrid Machine
Translation for the accurate implementation of Eng-Hindi
Translation when pronouns are used for gender recognition.

sitasings song

ससीततागसीतगतातसीहह

ram sings song

रतामगसीतगतातताहह

riyasings song he sings song

ररयतागसीतगतातसीहह
वहगतानतागतातताहह
वहगतानतागतातसीहह

she sings song

अजललिगसीतगतातसीहह


anjalisings song

वहगतानतागतातसीहह

she sings song
According to statistics sings->गतातसी
If input statement is
He sings song.
Output is: वहगतानतागतातसीहह !
A. Our Module
Input is: He sings song.
In our module if statement contains pronoun then statistical analysis is carried out for both pronoun and verb
According to statistics sings->गतातता

Figure 1: System Design.

2

Published By:
Blue Eyes Intelligence Engineering &
Sciences Publication Pvt. Ltd.

International Journal of Engineering and Advanced Technology (IJEAT)
ISSN: 2249 – 8958, Volume-X, Issue-X

sita sings song

ससीततागसीतगतातसीहह

ram sings song

रतामगसीतगतातताहह

riya sings song he sings song

|

ररयतागसीतगतातसीहह

Ekta Ahuja is a Computer Engineering student from VESIT, Mumbai.
She has made various projects using Python and believes that technology and management should go hand in hand
Karan Kashyap is a Computer Engineering student from VESIT,
Mumbai. He has made projects in ASP .NET and Java. He has keen interest in technology in general and looks at himself dealing with finance and analysis in related fields

वहगतानतागतातताहह
वहगतानतागतातसीहह

she sings song

अजललिगसीतगतातसीहह


anjali sings song

वहगतानतागतातसीहह

she sings song
If input statement is
He sings song.
Output is:
वहगतानतागतातताहह !
VI

SCOPE AND LIMITATION

Our module has been developed for the sentences of the form ‘Subject Object Verb’. It needs to be expanded to other sentence formats to improve efficiency and increase the usage scope. The significance and relevance of translation in our daily life is multidimensional and extensive. It is through translation we know about all the developments in communication and technology and keep abreast of the latest discoveries in the various fields of knowledge, and also have access through translation to the literature of several languages and to the different events happening in the world.
The future scope of our project at the major scale comes into significance when translation between English to Hindi is done using sentences of the type Subject-Object-Verb (in
English) where subject is name of the person or any pronoun that addresses a gender (he/she).
ACKNOWLEDGEMENT
We are grateful to Vivekanand Education Society’s
Institute of Technology for providing continuous support throughout the work.

REFERENCES
[1]

RAY SMITH, AN OVERVIEW OF THE TESSERACT OCR
ENGINE,PROC. NINTH INT. CONFERENCE ON DOCUMENT ANALYSIS
AND RECOGNITION (ICDAR), IEEE COMPUTER SOCIETY (2007),
PP. 629-633

[2]

KEVIN
KNIGHT,
A
STATISTICAL
MT
WORKBOOK,JHU SUMMER WORKSHOP, APRIL 30,1999

[3]

A Srivastav, English to Hindi Machine Translator Using GMT and RBMT Approach Proceedings of the Second International
Conference on Advances in Computing and Information
Technology (ACITY) July 13-15, 2012, Chennai, India

TUTORIAL

3

Published By:
Blue Eyes Intelligence Engineering &
Sciences Publication Pvt. Ltd.

Similar Documents

Free Essay

On Translation of Public Signs from Chinese to English

...etc. All these international practices make the public signs essential to all the foreign people, from government officials to entrepreneurs, from scientists to overseas students, from artists to sportsman. Under such a circumstance, public signs become a mirror reflecting the international image of China. The present paper has discussed the following aspects of C-E translation of public signs: the range and classification, the characteristics, issues in the course of translation, applicable principles and special requirements for translators for a better regulation of Chinese public signs into English. Generally speaking, the C-E translation of public signs is a challenging work, demanding a great deal of practice and broad knowledge about the English-speaking countries’ culture, custom, pattern of thought, etc. Huang Shouyi, the Secretary of the Translators Association of China and the vice president of the International Federation of Translators, pointed out that the publicity-oriented C-E translation should follow the principle of “three closeness”, specifically, to be close to the reality of the development of China, to be close to oversea receptors’ need, and to be close to the foreigners’ way of thinking. Therefore, much attention should be paid not only to linguistic competence but also to cross-cultural awareness. Nowadays, in a world of globalization, public signs are vital for international communication. Though a series of problems do exist, the official agencies...

Words: 381 - Pages: 2

Free Essay

Ethical Perspective

...Ethical Perspective of Sumer Institute of Linguistics Ethical Perspective of Summer Institute of Linguistics The Summer Institute if Linguistics (SIL) is a global organization that studies, develops, and documents languages from all over the world. The organization is a non-profit US based. This Christian organization promotes literacy in countries around the world focusing on the lesser and minority areas. SIL started out as a small summer program. Students would go around the world to assist missionaries. Missionaries travel to indigenous tribes and poor areas to teach the natives of that land about Jesus. Because the people they were ministering to were uneducated, usually they were unable to read and in many cases had no form of written languages. Because the need was so great, SIL became a year round organization and partnered with Wycliffe Bible Translators. SIL and Wycliffe members and missionaries are trained to live in areas of the world with limited resources. Basic living necessities are limited. The members and missionaries, after their training is complete moves to unknown parts of the world, with their family, and live amongst the very people they are trying to minister to. Without a doubt, this is no east task. In the United States, basic living requirements are taken for granted. Americans live in a land of abundance. When these members and missionaries move into their appointed lands, they are no longer afforded the same amenities they had while...

Words: 789 - Pages: 4

Free Essay

Ethics

...You are asked to translate and certify a document which is written in more than one language. You are able to read all the languages but are only accredited in one language. What should you do? I would inform the client that even though I’m able to read all the languages in the document, I am accredited only for one language and therefore I am not able to undertake the assignment. As a translator keeping the client aware of the NAATI level is very important for the code of ethics of a translator. Even though I am able to read all languages on the document I am accredited only for one language. And the other languages are not in my competence. As it is stated on the code of ethics competence is one of the most important aspects for a translator. 105 words You are asked to translate an Academic record. When the client arrives to collect the translation, they try to get you to change the results, claiming the “B” should have been an “A”. They insist you change it. What should you do? As a translator I will always be firm in my translations at all times. As a translator I need to abide to the code of ethics and it is not ethical to alter any of the content of the document. As it is stated on code of ethics Accuracy is one of the most important aspects of translation. As a translation it is not ethical to omit, alter or make any additions to the documents that I translate. In this basis if I am not able to omit or alter anything on the translation. However if it seems...

Words: 827 - Pages: 4

Premium Essay

Translated Narrative Analysis

...narratives too, with a much more emphasis on the target language and cultural system. The point is that if the original narratives, as we discussed above, can be ontological, interpersonal, communicative, social, fictional and ideological acts, so should be the translated narrative texts. And, if there happen some frequent changes in POV and FID, there may appear some changes in the ontological, ideological, and social functions of the text. In fact, the translated narratives are first and foremost, the reproductions, at the very broad sense of the word, of the originals in whatsoever form and content. They, however, have got something more:...

Words: 1097 - Pages: 5

Free Essay

Website Design

...has a very short attention span and if a website isn’t immediately engaging, the user will move on. There are also language barriers that prevent websites from being useful. Websites are created in every country, in every language, in every dialect of language. The content needs to be translated for the information to be universal. There are many tools and websites that translate information to be understandable to every internet user on the planet. Google Translate. Initial problems, with my last name being Bell, translation did not realize it was a surname, and translated it. I initially translated my page into German, where it translated it to Glocke. There are phrases that it also didn’t pick up and translate at all. Management Accounting and Bachelor of Science were not translated and left in English. However, when I used the site www.online-translator.com to translate my page to German, it realized that Bell was a surname and did not translate, and it also translated Management Accounting and Bachelor of Science properly into German. I would say that out of the two translators, the online-translator.com site performed better than the Google Translate. While machine and program ran translators are useful for basic language sites. But there are times when a professional translator will be required. Programs aren’t able to pick up on the nuances of every language. There are words that exist in German that have no...

Words: 1101 - Pages: 5

Premium Essay

Whatever

...In Other Words This book addresses the need for a systematic approach to the training of translators and provides an explicit syllabus which reflects some of the main intricacies involved in rendering a text from one language into another. It explores the relevance of some of the key areas of modern linguistic theory and illustrates how an understanding of these key areas can guide and inform at least some of the decisions that translators have to make. It draws on insights from current research in such areas as lexical studies, text linguistics and pragmatics to maintain a constant link between language, translation, and the social and cultural environment in which both language and translation operate. In Other Words examines various areas of language, ranging from the meaning of single words and expressions to grammatical categories and cultural contexts. Firmly grounded in modern linguistic theory, the book starts at a simple level and grows in complexity by widening its focus gradually. The author explains with clarity and precision the concepts and theoretical positions explored within each chapter and relates these to authentic examples of translated texts in a variety of languages, although a knowledge of English is all that is required to understand the examples presented. Each chapter ends with a series of practical exercises which provide the translator with an opportunity to test the relevance of the issues discussed. This combination of theoretical discussion and...

Words: 109520 - Pages: 439

Premium Essay

Relation Between Accounts and Finance

...About Microsoft Word 2010 | Use this guide to learn about some of the new and improved features which are available in Microsoft® Word 2010. To learn more about other new features in Microsoft Word 2010, go to http://office.microsoft.com/en-us/. Topics in this guide include:Understand Co-AuthoringUse the Navigation PaneUse Polished FindWork with Translation ToolsUnderstand Paste Live PreviewUse Improved Text EffectsIncorporate ScreenshotsWork with Additional Microsoft SmartArt® Graphics | Customization note: This document contains guidance and/or step-by-step installation instructions that can be reused, customized, or deleted entirely if they do not apply to your organization’s environment or installation scenarios. The text marked in red indicates either customization guidance or organization-specific variables. All of the red text in this document should either be deleted or replaced prior to distribution Understand Co-Authoring In Word 2010, multiple authors can now edit a document at the same time, while seeing who is working on other sections. If changes are made by others, new additions and authors are highlighted so you know what change was made where and by whom. Notes * Co-authoring features require Microsoft SharePoint® Server 2010 or Microsoft Windows Live™ Skydrive®. * Content synchronization is controlled by the user. Your changes are available to others when you save your changes. Others get your changes when they save their file. You must...

Words: 1518 - Pages: 7

Free Essay

Translating

...own language. When my daughter was learning to talk, she started extremely young. It started about when she was six months old. She would cry and then start saying “Oya, Oya” which took some time to decipher, but we quickly figured out that she would say oya when she was hungry. When I returned from Korea right around eighteen months, she started to develop short phrases. It wasn’t till she spent some time with her older cousin that her language skills jumped ahead of her vocabulary. She started speaking very fast, similar to a teenage girl on the phone. She would compact three or more words together in a tightly designed package. When people would hear her talk, my wife mainly would be her translator, even to me since I was gone for a year. My daughter would point and say what she was wanted; my wife would be able to translate what she was communicating. When I returned home from Korea, I was like a foreigner to my daughter’s version of English. I was always asking what she wanted and my wife had to play translator again. To the outside world my daughter spoke almost gibberish, but after spending some time with her, practically anyone could understand everything that she expressed. It has taken a lot of patience, work, and encouragement to get my daughter to slow down and speak normally. Now my son is a toddler and the translator job has started again, but in a different way. He speaks in short two or three word phases. This time it seems I am more adept...

Words: 416 - Pages: 2

Premium Essay

Wireless Headset Translator

...Wireless headset translator Two sources that were used in order to come up with our new invention were the ITA database and UNWTO (world tourism organization network). Taking a look at these sources, we found that the rate of tourism increases each year. To take advantage of this growing market, we came up with a new invention, which allows consumers to use it if ever they cannot speak the language of the foreign country there currently in. These sources made us realize how much money goes into travelling each year. On average, people in today’s society speak up to five different languages and with hundreds of different languages in the world; we believe that this product will be beneficial. As seen in the sources presented every yearly quartile accretion at least 3% of tourists in Canada alone. Our idea starts off with an earpiece that has a microphone attached to it. Similar to a Bluetooth “office headset”, our product will have a slim, lightweight design which is portable and easy to carry with you. The way this product works is very simple and effective. When the consumer speaks his language that he is most confortable with into the microphone, the headset picks up the signal and translates it to the set language. The way you set your language is very simple. The technology used in this product is no different then the Bluetooth technology that one would use to connect his/her phone to the car. Bluetooth technology has become and common amongst drivers who...

Words: 685 - Pages: 3

Free Essay

Health and Social Care Unit 1 P4

...Communication and interpersonal interaction * Staff training * Assessment of needs * Promoting rights * Confidentiality * Defusing aggression * Appropriate verbal/ non- verbal communication * Building relationships * Appropriate environment * Attitude and confidence Defusing aggression – a way to defuse aggression is by using assertion. Assertion is when a person can control their emotions that make them run away or fight back. To be assertive you would need to learn how to relax and stay calm in a tough situation. By building a relationship with a patient, a nurse could calm down a patient since she knows a lot about the patient therefore, he/she will know how to calm down the patient. A nurse would use non- verbal communication to calm down a patient also she would use verbal communication to speak with the patient and ask him/her what’s wrong. Assessment of needs – This will allow the nurses and the doctors to know what is best for the patient and how to deliver the best quality care. This strategy will make sure that the patient receives all the help that he needs and that the nurses know how to deliver the help. For example, if a patient cannot speak for some reason, the nurse could check the patients needs assessment and make sure that he is getting everything that he needs. For example, medicine. Confidentiality – This strategy is will allow the patient to feel more comfortable knowing that the nurses or the doctors won’t...

Words: 1401 - Pages: 6

Premium Essay

International Consideration for E-Commerce for Small Company

...insurmountable risks to deal with problematic international issues like cultural differences, language barriers, system infrastructures, legalities and finances. As a customer-centric company, the key to survival is customer satisfaction. Identifying cultural sensitivity issues to elicit customer satisfaction at the local level is challenging, because human behaviors are uniquely different; however to entertain this issue on an international front, requires hefty research and development budget. It is fallible to assume that people behaviors are all the same everywhere; therefore, the same customer service model can be used. Language barriers present another international impediment for a small business. According to Becker (2008), “…research have shown that customers are more likely to buy goods and services from Web sites in their native languages, even if they can speak and understand English well” (chapter 1.3, Background section, para. 1). Using language translation software on a Web site is not a viable option as human subtleties, colloquialism and idioms maybe lost in translation to the native reader. Language translators and Web designers who are experts come at a pricey rate. The system infrastructures to support languages are costly due to different encoding schemes. There is not a universal standard code for information interchange, which means that there is not one standardized language used by all end users. Another international consideration in e-commerce is the...

Words: 297 - Pages: 2

Premium Essay

Overcoming an Obstacle

...learning a second language. To be successful I established two specific few goals to help guide me in the right direction. First, and I feel the most important goal, was to speak the language at all times. The second goal I made was to read the language whenever possible. Speaking the language was the most important goal to me because the act of speaking words helped ingrain those words and their definitions into memory. I always carried a dictionary/translator with me so I could look up words I did not know. In the beginning, not having a proper core knowledge of the language made this goal extremely difficult. As time progressed, however, I realized that it was becoming easier and easier to not only speak, but I required less time to think of what I was trying to say. Reading the language was another important goal to help me learn a second language. Reading the language was very helpful for increasing my vocabulary. I would always read out loud while working on this goal, which in turn also helped me achieve the first goal, as well. As I did with speaking the language, I kept a dictionary close by as to look up the words I did not know. Again as time went on, I noticed that the frequency of having to look up words began to decrease. Having to learn a second language is a difficult task to accomplish. The setting of goals was crucial to the success of my accomplishment. Had I not made goals I probably would not have learned a second language, at least, not...

Words: 291 - Pages: 2

Premium Essay

Conflict Management

...require a lot more interaction and communication. This will give everyone the chance to meet face to face and becoming familiar with each other and this will be a fun time to come to a different city. Puerto Rico this time of the year is beautiful with temperatures in the 80’s. The efforts we make towards are team gathering I believe this will allow us to have an edge on teaching other teams that will be virtual teams and those teams that are forming in the future how virtual teams can work. On our agenda will be addressing the items that we all agreed on in our last group meeting. Please see below what will be addressed on the agenda. 1. Addressing members who were relocated to a different location 2. Communication Issue 3. Language Barrier 4. Planning and Following through 5. Ensuring Processes are complete 6. Ensuring backup when members are sick or on vacation One of our biggest challenges is the...

Words: 612 - Pages: 3

Premium Essay

Barriers to Communication in Health and Social Care Unit 1 Level 3

...clearly and slowly, and/or repeating, rephrasing what has been said, to help people understand what is being said to them. Time should be given to the message receiver; so that they can digest the information they have received and think about how they want to respond. Electronic devices can also be used, such as text phones, telephone amplifiers and hearing loops, and it is important to give the individuals using the devices enough time to use it whilst communicating. An induction loop system helps deaf people hear sounds more clearly by reducing or cutting out background noise. Conditions such as cerebal palsy, Down's syndrome, and autism tend to limit a person's ability to interpret other people's non-verbal communication, such as body language, and their ability to communicate verbally. In a care setting these needs should be attended to, so that effective communication can be achieved between the care worker and the individual with the condition. For example if...

Words: 882 - Pages: 4

Premium Essay

Health and Social Care Unit 1

...UNIT 201/ SHC 21 Q1) Identify the different reasons why people communicate? There are varies types of communication but all come under one of two categories. One being Verbal Commination and the other being Non-Verbal Communication. Verbal Communication is when one person communicates with another by using speech and Non-Verbal Communication is when one person communicates with another by using body language. People communicate with one another for many different reasons and for most this is an essential part of their daily lives. I have listed below a number of reasons of why I think people choose to communicate: * To express feelings * To teach * To learn * To connect * To share views and opinions * To take control * To compliment Q2) Explain how effective communication affects all aspects of working in adult social care settings? Effective communication is an extremely important part of working within adult social care. This is because we do not only have to communicate with one another as professionals but we also have to communicate with the people that we are here to support. * Clear & Concise - by having effective communication we are able to be clear and concise on what the needs of the individual are and how these needs can be met. * Mistakes - effective communicate also promotes the chances of less mistakes from happening by avoiding any inappropriate or unclear information being given or by something that has simply been...

Words: 3636 - Pages: 15