Voice Integration in Mobile Banking and other Banking Services. Shubham Dodia1, Umang Agrawal2, Dhananjay Pathak3 Prof. Sanjay Ghodke4, Mr. Saurabh Chande5 1,2,3,4Dept. of Computer Engineering MIT Academy Of Engineering, Pune, India 5Persistent Systems Ltd. Nagpur, India [email protected], [email protected],[email protected] ABSTRACT In today’s cloud-first mobile-first world, nearly everything is integrated on a smartphone. Every common man uses a smartphone today. The main reason behind this is the reduction in production cost of smartphones as well as increasing competition for new technologies amongst mobile vendors. Developers can easily exploit this situation to solve some real-life that we face in today’s world. The basic idea behind the project is come up with a system that supports voice interaction with the user for Mobile banking on a smartphone. User interacts and authenticates via voice commands with the application. The proposed application will run on Android, JAVA (for feature phones) and also a web version for other platforms. Our focus will be on User Interaction for the Banking Services. The system will provide responses to the questions posed as well as the commands issued by the user. The assistant will recognize the verbal input using Google Speech to Text API and Google Text to Speech API. Besides the Voice Actions API will serve the purpose of user defined voice commands for enhanced interaction with the user for banking transactions. The system will not only enhance the overall voice integrated mobile banking system domain but help the differently-abled to carry out the required transaction effectively. INDEX TERM: Voice Assistant, Mobile Banking, Banking for Blind, Google Voice Actions API, Google Text-to-Speech API, Google Speech-to-Text API.

I.

INTRODUCTION

People love their mobile phones. It is a device to keep in touch with the external words personally and professionally to make the life simpler. Initially, simple mobile phones were only used for talking and texting. We have witnessed the growth in the last decade in mobile technology where simple phones turned into Smart phones. These Smart phones are now loaded with breakthrough technologies available in the market such as high-end Cameras, inbuilt Music Players, Tablet PC’s and now coming up with TV features as well. Internet is readily available on these Smart phones through which people are staying connected with social, personal and professional life. To make this even more exciting, different applications are making their way into the life of Smart phones adding more interesting features to it. Still, people have to type in different messages and chats and it becomes clumsy task to type in such long texts time to time. In this world where technology is a key focus, what we have is even interesting speech technologies which help us to build this app: Speech Server. Using service component of phone and standardized communication protocols, we have come up with this Voice Message System. We are using the openly available and most widely used Android Operating System to publish this app. Speech recognition also

appeared as part of ongoing research in progress in 1950s, but was not so popular until the mid- 2000s. Initial versions of these Speech recognition technologies are now evolved to a great extent and are rapidly getting popular amongst the users. Siri is one of the most prominent examples of a mobile voice interface where latest iPhone have built in voice activated personal assistant. Similarly, Android, Windows, and other mobile platforms are drastically evolving with the Speech Recognition technologies through many applications. While these interfaces still have considerable constraint, we are inching closer to machine interfaces we can actually talk to. Let’s look into this application with a closer view on insights. II. LITERATURE SURVEY The research conducted so far in development of voice integration in mobile banking and other banking services are discussed in this section. The set of challenges outlined above span several domains of research and the majority of relevant work will be reviewed in this section. In this section, basic concept is discussed for better understanding of the project. [1] Asha Bharambe , Adwait Vyas , Vedant Pandit , Chinmay Pai , Sanjay Wadhwa, presented A HINDI LANGUAGE PERSONAL ASSISTANT FOR ANDROID in International Journal of Computer Engineering and Applications, Volume IX, Issue III, April 15 (ISSN 2321-3469). An Android-based Hindi language based personal assistant, which will understand colloquial Hindi and provide the necessary response. The system will provide responses to questions posed, as well to commands issued, by the user verbally in Hindi. [2] Kaveri Kamble, Ramesh Kagalkar presented Audio Visual Speech Synthesis and Speech Recognition for Hindi Language in (IJCSIT) Vol. 6 (2), 2015, 1779-1783. Their System developed for Hindi Text to Speech and Speech to Text Conversion mainly into the Hindi Language. TTS synthesizer will be helpful for Text Processing and Speech Generation. This system is very helpful for the people who are having hearing impairment and blind people [3] Surtar Shekhar, Kamad Neha presented Intelligent Voice Assistant Using Android Platform in IJARCSMS Volume 3, Issue 3, March 2015 (ISSN: 2327782 (Online)) This project is focusing on the Android development over the voice control (recognition, generate and analyze corresponding commands, intelligent responses automatically), Google products and relevant APIs (Google weather, Google search and etc.), Wikipedia API and mobile device references ranging from Speech-To-Text, TextTo-Speech technology, camera, messaging and other services and technology that are needed in daily life. [4] Renu Tarneja, Huma Khan, Prof. R. A. Agrawal Prof. Dinesh. D. Patil presented VOICE COMMANDS CONTROL RECOGNITION ANDROID APPS in IJERGS, Volume 3, Issue 2, March-April, 2015ISSN 20912730. The device proposed here is an interactive android smart phone, which is capable of recognizing spoken words. We propose to develop interactive application which can run on the tablet or any android based phone. The application helps the user to open any application as well as call any contact through voice. III. PROPOSED IDEA Integrate current mobile and web banking application to a voice assistant. Voice commands for various functions and services like:• • • •

Funds Transfer Utility Payment Checking Account Details Feedback and Queries

Goals:Voice based Mobile Banking Banking for the Blind Increased usability Notification for all banking events. Regional Support for India Focus on User Interaction

Figure 1: Block Diagram IV. PROPOSED SYSTEM

Figure 2: Architecture Diagram

The components of Voice Assistant are: 1) User The user will interact with the system by giving speech input in Hindi language. The system will process the input and will provide the appropriate response in Hindi language, if required. The response, if required will be given using Google's Text To Speech API. 2) Application The application is the vital part of the system. The user interacts with the system via application. The application is basically a software agent that collaborates with the user for performing tasks on user's behalf, at the same time hiding the task's complexity from the user. The application interacts with the Recognizer Service, Parser, and Android System to perform the task required by the user. 3) RecognizerService It is a Speech Recognizer class of android.speech package provided by the Android API. The application will send the speech input from the user in the form of recorded speech to the RecognizerService. The RecognizerService will transform the recorded speech in Hindi to Hindi text and send it to application. 4) Parser The parser will receive the Hindi text from RecognizerService via application. The function of parser is to extract object and action from Hindi text. The Hindi text is basically in the form of commands. A single command can be of various forms. The parser will forward the object and action pair to the Dialog Manager. 5) Dialog Manager Some commands may require additional parameters. The function of dialog manager is to check whether the required additional parameters for a particular command are present or not. If the required additional parameters are not present, then dialog manager will ask the user to provide additional parameters. Then, dialog manager will send object, action and additional parameters to the parser. 6) Android System The application receives object and action or object, action and additional parameters from the parser and depending upon the action, the application will call the Android system via system call. Android System is basically various applications like call, message, contact, Gmail, Hangouts etc. which are present on the android phone. These applications will perform the desired task and will return the result to application. The application will give output to the user and also provide responses in Hindi language, if required. V. ALGORITHM 1. Start the application 2. autheticateUser() 3. checkVoiceRecognition ()

//Check whether packages required for ACTION_RECOGNIZE_SPEECH are installed and

imported. 4. SpeakButtonClick() // Invokes onCLick() event for Voice Recognition 5. StartVoiceRecognitionActivity() //Handle voice Recognition Services Recognizer Intent

Create

new

Set RecognizerIntent language form to free form 6. onActivityResult() //Check if request code matches or error Populate the result in the array list Choose the best result 7. Get the Text query from the voice input 8. Send the text based query the bank 9. processTransaction() // Carry out the transaction 10. checkTransactionStatus() // Success or Failure 11. voiceResult() // TTS for Voice result 12. textResult() // Display text Result 13. Ask the user if he wants to carry out more transaction? 14. If (true) Goto step 3 Else Continue; 8. Stop the application VI. RELEVANT MATHEMATICS ASSOCIATED WITH THE PROJECT Input: 1. Voice Input from the user 2. Text based query Output: 1.

Voice Output from the Application

2. Text based result for Confirmation

Mathematical Model: S = {s, I,O, F,V,fi, fo} Where, S - Solution set, s - start state, I - Set of Input, O - Set of Output, F - Manipulation function, V – Transaction Function , fi - Input function(Input will be voice), fo - Output function Now, F is Manipulation function which performs various operations. F = {F1,F2,F3,F4,F5,F6,F7,F8} F1 - Authentication F2 - Voice Command Recording F3 - Regional Voice Recognition F4 - Conversion of voice to text F5 - Extraction of Object and Action F6 - Checking for necessary parameters

F7 - Transaction processing by bank F8 - Confirmation message by Voice Assistant Let suppose, V = {V1,V2,V3,V4} (Set of transactions) Where, V1= Request for Fund Transfer V2= Request for Utility Payment V3= Request for Account enquiry V4= Feedback or Queries V1 = {F1,F2,F3,F4,T1,T2,T3,F7,F8} V2 = {F1,F2,F3,F4,T1,T2,T4,F7,F8} V3 = {F1,F2,F3,F4,T1,T2,F7,F8} V4 = {F1,F2,F3,F4,F7,F8} Where, T1 = Connection to Bank Server T2 = Extract Account Details T3 = Transfer Funds T4 = Make Utility Payment VII. OUTCOME The outcome of the application will be a solution that will suffice the need of Voice Integration in the mobile banking domain. Users can easily interact with the application via voice commands and carry out the required banking transactions. There will be various types of transactions that a user can carry out like transfer, balance check and utility payment. The user thus can easily interact with the application and carry out the required transaction efficiently. VIII. CONCLUSION This paper discusses about voice guided mobile banking using a Voice Assistant. All the current banking services available in the mobile banking domain are to be integrated in the proposed application. Services include fund transfer, utility payment, account inquiry and feedback along with queries. The Google Text-to-Speech and Speech-to-text API will not only serve on the Android version of the system but will also easily interpret voice in the mobile web version The Application will not only help the differently able but also provide ease of access to the mobile banking users to carry out all the existing mobile banking services. IX. FUTURE WORK In the near future, we wish to focus on different platforms and implement the solution for various languages. Also, the scope of the application will be expanded to various finance domains in which the banking sector operates. Also in the future, shifting the focus towards a mobile optimized website instead of the mobile application will be an important future prospect of the solution.

X. APPLICATIONS AREA FOR PROJECT. Applications of the project are: [1] Banking sector [2] Voice enabled ATM’s 9. ACKNOWLEDGMENTS Our sincere thanks to all those who are guiding us in this project. Our sincere thanks to our guide and mentor:Prof. Sanjay Ghodke, Asst. Professor, COMP, MIT-AOE, Pune Our external Guide: Mr. Saurabh Chande, Persistent Systems Ltd. , Nagpur XI. REFERENCES [1]

[2] [3] [4] [5]

[6]

[7]

[8] [9] [10]

A HINDI LANGUAGE PERSONAL ASSISTANT FOR ANDROID, Asha Bharambe , Adwait Vyas , Vedant Pandit , Chinmay Pai , Sanjay Wadhwa . International Journal of Computer Engineering and Applications, Volume IX, Issue III, April 15 (ISSN 2321-3469) Audio Visual Speech Synthesis and Speech Recognition for Hindi Language Kaveri Kamble, Ramesh Kagalkar, (IJCSIT) Vol. 6 (2) , 2015, 1779-1783 Intelligent Voice Assistant Using Android Platform, Surtar Shekhar, Kamad Neha, IJARCSMS Volume 3, Issue 3, March 2015 (ISSN: 2327782 (Online)) VOICE COMMANDS CONTROL RECOGNITION ANDROID APPS Renu Tarneja, Huma Khan, Prof. R. A. Agrawal Prof. Dinesh. D. Patil, IJERGS, Volume 3, Issue 2, March-April, 2015ISSN 2091-2730 A Hindi Speech Actuated Computer Interface for Web Search,kamlesh sharma ,Dr. S.V.A.V. Prasad,Dr. T. V. Prasad,(IJACSA) International Journal of Advanced Computer Science and Applications Vol. 3, No. 10, 2012 Translation of Text to Speech Conversion for Hindi Language Kaveri Kamble, Ramesh Kagalkar, International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Impact Factor (2012): 3.358 Volume 3 Issue 11, November 2014 www.ijsr.net Designing Personal Assistant Software for Task Management using Semantic Web Technologies and Knowledge Databases, Purushotham Botla, MASSACHUSETTS INSTITUTE OF TECHNOLOGY JUNE 2013 Google's Text To Speech API http://developer.android.com/reference/android/speech/tts/TextToSpeech.html android.speech package by Android API http://developer.android.com/reference/android/speech/SpeechRecognizer.html Cortana ,from Wikipedia,the free encyclopedia http://en.wikipedia.org/wiki/Cortana