Voice control of smart home by using Google Cloud Speech-To-Text API

Abstract

This work explains the process of creation two applications. The first application is an Android application that uses Speech-To-Text transcription by implementation of web service Google Speech API and Bluetooth communication as the client. The second application is a Windows application written in C# language using Bluetooth web service as the server. The applications show how Google Speech API works and how to implement these web services. The thesis consists of three main parts: Google Speech API, Bluetooth communication between the two devices, and the method to build an application using these features. The first part describes various ways to convert Speech into text form using Google Speech API and its features. It also explains what Speech to text conversion means and how this technique works. The second part describes the principle of operation of Bluetooth communication; how this communication works; how to implement Bluetooth service to a Windows application as the server and to an Android application as the client side. It also introduces its benefits but also its restrictions and its shortcomings. The third part deals with creating of the Windows and Android applications. This part describes how to implement these services to the specific applications, what is needed to set and how these applications work. The goal of the thesis has been met, both applications were created successfully and the whole system works correctly.