DocumentCode :
2138893
Title :
A system for automatically annotating traditional Irish music field recordings
Author :
Duggan, Bryan ; O´Shea, Brendan ; Cunningham, Padraig
Author_Institution :
Sch. of Comput., Dublin Inst. of Technol., Dublin
fYear :
2008
fDate :
18-20 June 2008
Firstpage :
25
Lastpage :
32
Abstract :
This paper presents MATT2 (machine annotation of traditional tunes). MATT2 is a novel system which can automatically annotate field recordings of traditional Irish music with useful metadata such as tune name, key signature, time signature, composer and discography. MATT2 works by using a number of algorithms to automatically transcribe digital audio to be annotated to the ABC music notation language. It then compares these transcriptions against a corpus of 860 human made transcriptions in ABC using a variation of the edit distance algorithm. Results using MATT2 to annotate fifty recordings of flute and fiddle tunes demonstrate a high success rate at annotating recordings made by different musicians. Additionally, several of the recordings successfully annotated in testing MATT2 were recorded in imperfect conditions, with badly degraded audio.
Keywords :
audio signal processing; information retrieval; meta data; music; ABC music notation language; MATT2; automatic annotation; digital audio; machine annotation; traditional Irish music field recordings; traditional tunes; Audio recording; Books; Digital signal processing; Disk recording; Humans; Internet; Multiple signal classification; Music information retrieval; Signal processing algorithms; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Content-Based Multimedia Indexing, 2008. CBMI 2008. International Workshop on
Conference_Location :
London
Print_ISBN :
978-1-4244-2043-8
Electronic_ISBN :
978-1-4244-2044-5
Type :
conf
DOI :
10.1109/CBMI.2008.4564923
Filename :
4564923
Link To Document :
بازگشت