Audio data classification by means of new algorithms

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F13%3A00204936" target="_blank" >RIV/62156489:43110/13:00204936 - isvavai.cz</a>

Výsledek na webu

<a href="http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6613984" target="_blank" >http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6613984</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/TSP.2013.6613984" target="_blank" >10.1109/TSP.2013.6613984</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Audio data classification by means of new algorithms

Popis výsledku v původním jazyce

This paper describes classification of sound recordings based on their audio features. This is useful for querying large datasets, searching for recordings with some desired content. We use musical recordings as well as birdsongs recordings, which usually have rich structure and contain a lot of patterns suitable for classification. We present two different classification methods, one for musical recordings and one for birdsongs. These methods are compared and their differences are discussed. We use feature vectors that capture the audio content of recording as a whole piece and then classify these feature vectors using combination of the Self-organizing map and the Learning Vector Quantization, which represent a powerful algorithm using unlabeled as well as labeled data. In case of birdsongs we use feature vectors representing time frames of a recording.

Název v anglickém jazyce

Audio data classification by means of new algorithms

Popis výsledku anglicky

This paper describes classification of sound recordings based on their audio features. This is useful for querying large datasets, searching for recordings with some desired content. We use musical recordings as well as birdsongs recordings, which usually have rich structure and contain a lot of patterns suitable for classification. We present two different classification methods, one for musical recordings and one for birdsongs. These methods are compared and their differences are discussed. We use feature vectors that capture the audio content of recording as a whole piece and then classify these feature vectors using combination of the Self-organizing map and the Learning Vector Quantization, which represent a powerful algorithm using unlabeled as well as labeled data. In case of birdsongs we use feature vectors representing time frames of a recording.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2013

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the 36 th International Conference on Telecommunikations and Signal Processing

ISBN

978-1-4799-0404-4

ISSN

—

e-ISSN

—

Počet stran výsledku

5

Strana od-do

507-511

Název nakladatele

TSP

Místo vydání

Rome, Italy

Místo konání akce

Rome, Italy

Datum konání akce

2. 7. 2013

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—