|
|
Voice recognition and text conversions using Speech Recognition JS API
NOVÁK, Martin
The point of the bachelor's thesis is to process the issue of voice recognition and voice input using the JavaScript interface SpeechRecognition. This interface is part of the Web Speech API, which allows you to integrate voice data into web applications and contains two components - text-to-speech (SpeechSynthesis) and asynchronous speech recognition (SpeechRecognition). JS API technology allows you to recognize the voice coming into the input, then convert it to text, which can then be used in many ways, such as simply typing as a dictaphone, or invoking other functions using a word command. A very positive and especially current feature is the fact that it is a so-called blind friendly technique, ie technology designed for visually impaired users of web applications in this case. In the theoretical part, the author will focus on the introduction of basic web technologies related to the issue with a focus on JavaScript and especially on the SpeechRecognition API itself and its implementation and incorporation using the aforementioned JavaScript. In the practical part, a web application will be created, which will demonstrate the possibilities, features and advan- tages or disadvantages that this voice recognition technology currently brings.
|
guest ::
