Mogućnosti prepoznavanja govora korištenjem biblioteke SpeechRecognition

master's thesis

• Title (english): Speech recognition capabilities using the SpeechRecognition library
• Author: Margarita Tolja
• Mentor: Goran Igaly (mentor)
• Committee member: Goran Igaly (predsjednik povjerenstva)
• Committee member: Luka Grubišić (član povjerenstva)
• Committee member: Ilja Gogić (član povjerenstva)
• Committee member: Juraj Šiftar (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2022-03-04, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

U ovom radu predstavljena je biblioteka SpeechRecognition – popularna Python biblioteka za prepoznavanje govora. Na početku je dan povijesni pregled razvoja sustava za automatsko prepoznavanje govora, a posebno su istaknuta dva tipa najnovijih sustava. Nadalje, detaljno je opisana funkcionalnost biblioteke te njeno trenutno stanje razvoja. Integracija biblioteke je demonstrirana na primjeru vlastite desktop aplikacije za kreiranje glasovnih bilješki. Biblioteka se pokazala izrazito fleksibilnom i jednostavnom za korištenje, ali sa sobom nosi i određena ograničenja.

Abstract (english)

This thesis presents the SpeechRecognition library – a popular Python library for speech to text conversion. By way of an introduction, an historical overview of development of automatic speech recognition systems is given, with two types of the latest systems highlighted. Furthermore, the functionality of the library and its current state of development are described in detail. The integration of the library is demonstrated with a new desktop application for creating voice notes. The library proved to be extremely flexible and easy to use, but it also has certain limitations.

• Keywords: biblioteka SpeechRecognition; Python biblioteka; automatsko prepoznavanje govora
• Keywords (english): SpeechRecognition library; Python library; automatic speech recognition systems
• Language: croatian
• URN:NBN: urn:nbn:hr:217:805882
• Study programme: Title: Computer Science and Mathematics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra računarstva i matematike (magistar/magistra računarstva i matematike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2022-03-31 12:56:09