Pal: personal assistant system using low-cost computer

Automatic Speech Recognition (ASR) describes the ability of a computer to capture, identify and recognize the variety of human speech. It has been applied in many technologies such as personal assistant systems. Unfortunately, many personal assistant systems has been built in a way that may not always be disability-friendly and this causes the affected users whom are blind, disabled, illiterates and those who have physical limitations unable to enjoy the benefits of operating a computer. Hence, PAL is introduced. PAL is a personal assistant system built using low-cost device called the Raspberry Pi and open source voice-controlled software called Jasper. The functionalities of PAL includes searching for information on the internet, check for unread emails, schedule events in the calendar, manage a to-do list and translate texts through voice commands. Apart from that, a friendly graphical user interface (GUI) is also designed to display the output of each of the functional modules. Lastly, a number of tests are conducted to evaluate the performance and accuracy of the functional modules, GUI output display as well as response rate of the system. These tests include GUI output display test, user acceptance testing, the Command Success Rate (CSR) and Word Error Rate (WER) tests as well as response rate test. With the development of this project, it is hoped that PAL will be able to provide users with the benefits of using a computer in a more convenient and cost efficient manner