Systém syntézy řeči založený na optickém rozpoznávání znaků v programovém prostředí LabVIEW

Abstract

The bachelor thesis deals with the creation of a complete chain of written word to speech in the LabVIEW programming environment. The theoretical part discusses the partial steps of the entire TTS chain, i.e. both text recognition OCR and subsequent speech synthesis. The practical solution to the problem is divided into two separate applications. The first application is designed as an offline solution that uses the available LabView technology on the user's computer. The NI Vision Development Module is used for OCR. It is using text recognition according to the template. The SpeechSynthesizer system class is used for TTS. The second application is designed as an online solution using Google cloud services. In this case, the application is based on the REST API http requests, where it uses Vision and Text-to-Speech cloud modules of Google.