Original title: Detekce Akustického Prostředí z Řeči
Translated title: Acoustic Scene Classification from Speech
Authors: Dobrotka, Matúš ; Glembek, Ondřej (referee) ; Matějka, Pavel (advisor)
Document type: Master’s theses
Year: 2018
Language: cze
Publisher: Vysoké učení technické v Brně. Fakulta informačních technologií
Abstract: [cze] [eng]
Téma tejto diplomovej práce je klasifikácia audio nahrávky do 15 tried akustických prostredí, v ktorých sa ľudia bežne nachádzajú. Práca popisuje 2 metódy založené na GMM a i-vektoroch a ich vzájomnú fúziu. Na dátach zo súťaže DCASE dosiahol najlepší GMM systém úspešnosť 60.4% a i-vektor systém 68.4%. Fúzia GMM systému a najlepšieho i-vektor systému výsledok ešte zlepšila na 69.3%, čo by v dobe súťaže stačilo na 20. miesto z 98 odovzdaných systémov z celého sveta. The topic of this thesis is an audio recording classification with 15 different acoustic scene classes that represent common scenes and places where people are situated on a regular basis. The thesis describes 2 approaches based on GMM and i-vectors and a fusion of the both approaches. The score of the best GMM system which was evaluated on the evaluation dataset of the DCASE Challenge is 60.4%. The best i-vector system's score is 68.4%. The fusion of the GMM system and the best i-vector system achieves score of 69.3%, which would lead to the 20th place in the all systems ranking of the DCASE 2017 Challenge (among 98 submitted systems from all over the world).
Keywords: Acoustic scene classification; DCASE Challenge; filter banks; fusion; GMM; i-vector; i-vector extractor; linear Gaussian classifier.; MFCC; banky filtrov; DCASE Challenge; Detekcia akustického prostredia; fúzia; GMM; i-vektor; i-vektor extraktor; lineárny Gaussovský klasifikátor.; MFCC koeficienty

Institution: Brno University of Technology (web)
Document availability information: Fulltext is available in the Brno University of Technology Digital Library.
Original record: http://hdl.handle.net/11012/84928

Permalink: http://www.nusl.cz/ntk/nusl-385945


The record appears in these collections:
Universities and colleges > Public universities > Brno University of Technology
Academic theses (ETDs) > Master’s theses