Language Agnostic Chatbot

Language Agnostic Chatbot

M. Bharath, E. Vijay Prabhakaran, S. Subeesh, S. Senthilkumar

Department of Information Technology

M. P. Nachimuthu M. Jaganathan Engineering College Erode, India

Abstract – Chatbots have become an important part of modern digital systems by enabling automated interaction between users and computers. Most existing chatbot systems depend on cloud infrastructure and require continuous internet connectivity. This dependency creates privacy concerns and usage limitations. In this paper, a language agnostic chatbot is proposed which operates completely on a local system. The chatbot supports multilingual communication and allows users to interact through text, voice, and image inputs. A simple memory mechanism is also included to maintain conversational context during interaction. Since the system runs locally, it improves data privacy, reduces dependency on external servers, and provides faster response time. The proposed system demonstrates a practical approach for building secure and flexible conversational AI systems.

Keywords – Chatbot, Natural Language Processing, Multilingual Chatbot, Artificial Intelligence, Flask Framework

1. INTRODUCTION

   Artificial Intelligence has significantly improved the way humans interact with computer systems. One of the most widely used applications of artificial intelligence is the chatbot. Chatbots are computer programs that simulate human conversation and provide automated responses to user queries. They are widely used in areas such as education, customer support, and information services.

   However, most existing chatbot systems rely on cloud platforms for processing user queries. This requires continuous internet connectivity and may lead to privacy issues because user data is processed on external servers. In addition, many chatbot systems support only a limited number of languages.

   To overcome these limitations, this paper presents a language agnostic chatbot that can operate on a local system. The proposed system supports multiple languages and allows interaction through text, voice, and image inputs. The chatbot processes user queries using natural language processing techniques and generates responses using a locally running language model. By operating locally, the system ensures better privacy, faster response time, and unrestricted usage.

2. LITERATURE REVIEW
   1. Several studies have been conducted in the field of conversational AI and chatbot systems. Transformer based models have significantly improved language understanding in natural language processing tasks. These models help systems understand the context of user queries and generate meaningful responses.
   2. Large language models have also improved conversational capabilities of chatbots. However, many of these systems depend on cloud based infrastructure which requires internet connectivity and raises privacy concerns.
   3. Research on multilingual chatbot systems has attempted to improve accessibility for users from different linguistic backgrounds. Voice assistant technologies have also enhanced user interaction by allowing speech based communication with machines.
   4. Although these systems have improved chatbot functionality, many of them still depend heavily on internet connectivity and lack privacy protection. These limitations motivated the development of the proposed language agnostic chatbot system.
3. PROPOSED SYSTEM

   The proposed system is a language agnostic chatbot designed to run on a local machine. The system allows users to interact with the chatbot using text, voice, or image inputs. The chatbot processes user input using natural language processing techniques and generates appropriate responses.

   The system is implemented using Python and Flask framework to create the backend application. A local language model is used to generate conversational responses. Voice input is converted into text before processing, while image input is analyzed to provide relevant information.

   The chatbot also includes a memory component which stores previous interactions. This helps maintain context during conversation and improves the quality of responses. Since the system runs locally, it reduces dependency on cloud services and ensures better privacy for users.

4. SYSTEM ARCHITECTURE

   The architecture of the proposed chatbot system consists of several modules that work together to process user queries. The user interacts with the chatbot through a user interface which accepts text, voice, or image input.

   The input is sent to the backend server developed using Flask. The language processing module analyzes the user query and determines the appropriate response. The processed input is then sent to the local language model which generates the response.

   The system also includes a memory module that stores previous conversations to maintain context. Finally, the generated response is sent back to the user interface where it is displayed to the user.

   The architecture of the proposed language agnostic chatbot system is shown in Fig. 1.

   Since the chatbot operates locally, the response time was observed to be faster compared to cloud based chatbot systems. The results indicate that the proposed system provides a practical and secure conversational AI solution.

   Fig. 2 Chatbot Response for Text Query

   Fig. 3 Multilingual Chatbot Response

   Fig. 1 Architecture of the Proposed Language Agnostic Chatbot

5. RESULTS AND DISCUSSION

   The chatbot system was tested using different types of user inputs including text queries, multilingual queries, and image based inputs. The chatbot responses are illustrated in Fig. 2, Fig. 3, and Fig. 4. The system successfully generated relevant responses for user questions.

   The chatbot was also able to process multilingual input and provide appropriate responses. Image input testing demonstrated that the system could analyze uploaded images and generate advisory responses based on the image content.

   Fig. 4 Image Input Analysis by Chatbot

6. CONCLUSION

This paper presented a language agnostic chatbot capable of operating on a local system. The system supports multilingual interaction and allows users to communicate using text, voice, and image inputs. The use of a local language model improves privacy and reduces dependency on external cloud services.

The results demonstrate that the proposed chatbot can provide efficient responses while maintaining user privacy. The system can be further enhanced by integrating more advanced natural language models and expanding support for additional languages.

REFERENCES

1. A. Vaswani et al., Attention Is All You Need, Advances

   in Neural Information Processing Systems, 2017.

2. T. Brown et al., Language Models are Few Shot Learners, Advances in Neural Information Processing Systems, 2020.
3. S. Kumar et al., Multilingual Chatbot Systems for Conversational Applications, International Journal of Computer Applications, 2019.
4. Y. Li et al., Voice Assistant Technologies for Human Computer Interaction, IEEE Access, 2021.
5. J. Smith et al., Image Based Artificial Intelligence Systems for Visual Undertanding, Journal of Artificial Intelligence Research, 2022.