1. Introduction “There must be a language that doesn't depend on words, the boy thought. I've already had that experience with my sheep, and now it's happening with people.”[1] -Paulo Coelho “The Alchemist” These words by the author speak of a universal language understood by all the living beings –men and animals alike. Though, such a language may have existed –we have for long forgotten its usage. Today, even with the internet bringing the people together to make a global village, there still exists a barrier between people called –“Language”. There are numerous email services, chat messengers, voice and video conferencing tools existing today, however they are useful only when the medium of communication- the language- is same at all ends. Else, they are of little use. To overcome this last hurdle in bringing the people from various geographies, cultures, linguistics together-we planned to build a chat messenger which enabled people to send and receive messages in their own languages irrespective of what language the person at other end was using.

1.1 Review of the present scenario and the proposed solution The traditional and unchanged method of conversation between the people speaking different languages is to engage a translator between them. But, it is not always feasible and possible to deploy human translators for all the conversations. The advancements of computing and machine learning have led to the development of computer based translators. Though they solve the former problem to a large extent, their application in real time communication is hindered by following short comes:  Standalone applications for translation like IMtranslator. Not integrated with messengers and copy-paste procedures are not feasible for instant messaging or any sort of real time communication.  Add-ons and plug-ins like Microsoft TBOT which support only specific messengers like Windows Live messenger.  Other paid services with support for only few international languages. No service available for Indian languages like Hindi.  Even with few applications providing translation services, interfaces still in English.

1.1.1 The proposed solution: To overcome the above problems, we aim at building the Cross Language Insta-Translate Messenger which is an open source instant messenger just like Yahoo Messenger or AIM with facilities such as one to one text messaging, conference rooms, emoticons etc. but unlike them has some unique features like:  Translation built into it using the Google Translation API [2] which enables a very convenient and easy way for text chatting between people of different languages.  It also provides the interface in the language, the user desires.

1.2 Application fields of the proposed solution The proposed Cross Language Insta-Translate Messenger can be of great use in many day to day applications. The application fields range from hospitality industry to corporate world. We see few examples of its usage as below:  A hotel can install this application in all its rooms and kiosks and maintain a central customer support server. Whenever a tourist speaking a different language wants to know some information, he can easily chat with the customer service desk in his own language irrespective of whether the people at the desk don’t know his language or not.  In a software development company, like the one in which we did our project, there is many a times need of collaboration between people from various geographies like Europe and East Asian countries like China, Korea, Japan etc.. Most of the times communicating with them can become a problem as not very comfortable with English. Our solution is very useful in solving this problem.  Apart from these specific applications, there can many useful applications that may spring up in our daily life-like a – grandparent in Uttar Pradesh can chat using this solution with his grandson in USA who doesn’t know Hindi. Or it may even be a simple time-pass chat between a Serb and an Arab which may lead to a valuable cultural exchange.  Thus with our proposed solution supporting many languages like French, German, Finnish, Czech, Serbian etc. and our own Indian language Hindi, the applications are bound to be many.

2.Key concepts:

This section describes the key technologies used for development:

The Server The server's main function is to wait for incoming requests, and to service them when they come in. So the code to implement the server can be further broken down to the following steps:

1. Establish a server that monitors a particular port. This is done by creating an instance of the ServerSocket class. There are four different ways to create an instance of ServerSocket. They are: a. ServerSocket(), which simply sets the implementation that means everything is taken as default values. b. ServerSocket(int port), which creates a server-side socket and binds the socket to the given port number. c. ServerSocket(int port, int backlog), which not only binds the created socket to the port but also create a queue of length specified by the number passed as the backlog parameter. d. ServerSocket(int port, int backlog, InetAddress bindAddr), which creates a server-side socket that is bound to the specified port number with the queue of length specified by the backlog and bound to the address specified by the bindAddr argument. So to create a socket bound to port number 8888 with a backlog queue of size 5 and bound with address of local host the statement would be: ServerSocket server = new ServerSocket(8888, 5, InetAddress.getLocalHost() ); One point to keep in mind is that the above mentioned constructors return TCP sockets and not UDP sockets.
2. The next step is to tell the newly created server socket to listen indefinitely and accept incoming requests. This is done by using the accept() method of ServerSocket class. When a request comes, accept() returns a Socket object representing the connection. In code it would be: Socket incoming = server.accept();
3. Communicating with the socket, which means reading from and writing to the Socket object. To communicate with a Socket object, two tasks have to be performed. First the Input and Output stream corresponding to the Socket object has to be obtained. That can be done by using the getInputStream() and getOutputStream() methods of Socket class.
4. Once the client breaks the connection or stops sending the request, the Socket object representing the client has to be closed. This can be done by calling close() method on the Socket object.

The Client The main purpose of the client is to connect to the server and communicate with it using the connection. So coding a client requires the following steps:

1. Connect to the server. Connecting to the server can be accomplished in two steps: a. Creating a Socket object. The socket at client side just needs to know the host name (the name of the machine where server is running) and the port where the server is listening. To create a Socket object, there are seven constructors provided by the Socket class, of which the most commonly used are:  Socket(), which creates a new Socket instance without connecting to host.  Socket(InetAddress address, int port), which creates a new Socket object and connects to the port specified at the given address.  Socket(java.lang.String host, int port), which works the same way as Socket(), except that instead of an address, the host name is used. So to create a Socket object that connects to 'localhost' at 8888, the statement would be: Socket s=new Socket("localhost",8888);

b. Connecting to the server comes into picture if no argument constructor is used. It takes the object of the SocketAddress object as an argument. So to connect to localhost at port 8888, the code would be: Socket s= new Socket(); s.connect(new SocketAddress("localhost",8888));

2. Communicating with the server. Communicating with the server using a socket at client side is no different from how the server communicates with the client.

2.3.2 Message serialization [9] Serialization is the process of taking objects and converting their state information into a form that can be stored or transported. The basic idea of serialization is that an object writes its current state, usually indicated by the value of its member variables, to persistent storage. Later, the object can be re-created by reading, or deserializing, the object's state from the storage. Serialization handles all the details of object pointers and circular object references that are used when you serialize an object. In the message-queuing feature, serialization refers specifically to the process of converting an object or set of data into a message that can be sent to a queue, and then converting messages retrieved from the queue back into objects or data that your application can process. In our application, we specifically focus on converting message objects into bytes and back.

2.3.3 Google Translation API [10] Google Translate is as a beta service provided by Google Inc. to translate a section of text, or a webpage, into another language, with limits to the number of paragraphs, or range of technical terms, translated. Google Translate, like other automatic translation tools, has its limitations. While it can help the reader to understand the general content of a foreign language text, it does not always deliver accurate translations. Some languages produce better results than others. It does not apply grammatical rules, since its algorithms are based on statistical analysis rather than traditional rule-based analysis. Google translate is based on an approach called statistical machine translation, and more specifically, on research by Franz-Josef Och who won the DARPA contest for speed machine translation in 2003. According to Och, a solid base for developing a usable statistical machine translation system for a new pair of languages from scratch, would consist in having a bilingual text corpus (or parallel collection) of more than a million words and two monolingual corpora of each more than a billion words. Statistical models from this data are then used to translate between those languages. To acquire this huge amount of linguistic data, Google used United Nations documents. The same document is normally available in all six official UN languages (Arabic, Chinese, English, French, Russian, Spanish), thus Google now has a 7-language corpus of 20 billion words' worth of human translations. The availability of Arabic and Chinese as official UN languages is probably one of the reasons why Google Translate initially focused on the development of translation between English and those languages, and not, for example, Japanese and German, which are not official languages at the UN.

3.SYSTEM REQUIREMENTS :

3.1.1 Hardware Requirements Processor: 1GHz or higher Hard Drive: 1GB free disk space Memory: 1GB RAM 3.1.2 Software Requirements Operating system: Windows XP Service Pack 2 or higher with Regional Language Packs installed for inputting Local language characters

Java Runtime Envirmonment