Student Name

Rabeea SaleemFollow

Degree

Master of Science in Data Science

Department

Department of Computer Science

Faculty/ School

School of Mathematics and Computer Science (SMCS)

Date of Submission

Spring 2025

Supervisor

Dr. Sajjad Haider, Professor, Department of Computer Science, Institute of Business Administration, Karachi

Keywords

Urdu Voice Agent, Speech Recognition, Natural Language Understanding

Abstract

This project presents an intelligent Urdu voice system designed to process natural spoken queries from users, interpret their intent, and manage restaurant-related tasks such as order placement, order modification, booking reservations, and customer detail retrieval. The system makes use of a speech recognition module to transcribe Urdu audio, machine translation to convert between Urdu and English, and a modular AI pipeline for intent classification, information extraction, and response generation. It also integrates MongoDB in order to store information regarding customer orders and bookings. The backend functionality is built with Python, while Gradio has been used as a front-end tool. All in all, the application offers a conversational, voice-based interface capable of handling structured human to computer natural interactions, making it suitable for use cases such as restaurant assistants or customer service bots for Urdu-speaking regions.

Document Type

Restricted Access

Submission Type

Research Project

Poster.png (1322 kB)

The full text of this document is only accessible to authorized users.

Share

COinS