Voice and image recognition for smart pet app

Lee, Yen Long (2019) Voice and image recognition for smart pet app. Final Year Project, UTAR.

Preview

Abstract

Nowadays, there are many people feel depressed and lonely. The propose of this project is to develop a Smart Pet Application which is an Android application that can interact with people to reduce their loneliness. There are two parts in the Smart Pet Application: 1. Game design and logics. 2. Artificial intelligence (AI) engine to perform voice and image recognition. This report will focus on second part, which is descript how the AI engine work. In the propose system, TensorFlow Sequence-to-Sequence (tf-seq2seq) Model is used to be the machine learning model to let the Smart Pet can communicate with its owner. tf-seq2seq is a general-purpose encoder-decoder framework for TensorFlow that suitable for Conversational Modelling. On the other hand, Baidu Cloud face recognition SDK is used for the system to let the Smart Pet perform face recognition in order to recognize its owner. Baidu Cloud face recognition is a component on Baidu Cloud Computing Platform that suitable for real-time object recognition. We will combine the game design and logics together with the AI engine to become a complete Smart Pet Application.

Item Type:	Final Year Project / Dissertation / Thesis (Final Year Project)
Subjects:	Q Science > Q Science (General)
Divisions:	Faculty of Information and Communication Technology > Bachelor of Computer Science (Honours)
Depositing User:	ML Main Library
Date Deposited:	20 Aug 2019 12:12
Last Modified:	20 Aug 2019 12:12
URI:	http://eprints.utar.edu.my/id/eprint/3496

Actions (login required)

View Item