Digital intelligent virtual assistant (DIVA) with natural speech and accent recognition
J. Sayed, M. W. Ashour · IET conference proceedings. · 2022
A Virtual Assistant (VA) system is a computer program that recognizes designated language voice commands and executes tasks on the user's behalf. This paper aims to conduct a state-of-the-art review for Digital Intelligent Virtual Assistant (DIVA) and their extended use for diverse users. Furthermore, the paper proposes a workflow scheme which takes into account accent recognition, voice registration and gives more manual control to the users. The proposed design considers the user ease and their struggles with technology; hence the proposed system is destined to serve as a mobile application. The VA encompasses techniques such as Natural Language Processing, Speech Pattern Recognition, Machine Learning and Language Data Statistical Analysis to accomplish the aforementioned user goals. Subsequently, this research will indulge in data analysis using questionnaires filled out anonymously and the results of this survey will allow us to validate our hypothesis. The analysis results are expected to demonstrate how users view their Digital Virtual Assistants and address their concerns regarding security, accent support, control and ease of use. Finally the paper summarizes the findings, methodology and results obtained through questionnaires and highlighting the potential for future works.