Hybrid Multi Purpose Voice Assistant
Deborshi Lahiri, Purna Chandra Rao Kandimalla, A. Jeysekar · 2023
In this modern world everyone is busy in their works and there are many people around us who feel it very hard to organize their tasks, so everyone is looking for an intelligent assistant which can organize their tasks and perform some simple tasks with just a single command using voice assistants like Siri, Alexa, Google Assistant, and Bixby. Similarly, a chatbot is a computer program that uses natural language processing and machine learning to simulate or mimic a conversation with human users. Chatbots can also be used for a variety of purposes, including customer services, marketing, and providing information. They can also be programmed to perform specific tasks such as making appointments, placing orders, or providing recommendations. The aim of the paper is to propose a hybrid system which combines the traditional voice assistance system with a chatbot. This is not only combining both the advantages of chatbots and voice assistants but also gives us the option of customize the software to users based on their needs. This Hybrid Multipurpose Voice Assistant uses Natural Language Processing (NLP) to convert voice commands to instructions that will be sent to the voice assistant based on which it will perform the tasks. It uses Long Short-term Memory to process the instructions. Long short-term memory (LSTM) is a type of recurrent neural network (RNN) which is capable of learning and maintaining a record of information for long periods of time. Unlike traditional RNNs, which have a tendency to lose information over longer sequences, LSTMs use a special type of memory cell called a “memory gate” to control the flow of information into and out of the network. This allows LSTMs to selectively choose which information to keep and which to discard, making them well-suited for tasks such as language translation and speech recognition. The proposed Hybrid Multi-Purpose voice assistant will be a cost effective and efficient system which will understand and respond to a user's requests better and the interactions will be more human like. This project will also aim at helping people with a variety of disabilities and will be a step forward in the direction of making people, who suffer from different types of disabilities, lives easier. This voice assistant can be run both locally on a person's computer or externally using servers based on the requirement as the program is scalable and can work with existing technology and architecture.