Files in this item

FilesDescriptionFormat

application/pdf

application/pdfECE499-sp2017-walder.pdf (1MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Methods of onset detection for the bowed violin and why they failed
Author(s):Walder, Samuel
Contributor(s):Haken, Lippold
Subject(s):signal processing
music
onset detection
Abstract:An onset in musical signals is the time at which a note begins. Onset detection is important in computer analyses of musical performances in order to determine tempo, rhythm, and correctness. In this research, both existing and novel methods of onset detection on a dataset of 200 violin onsets played in various styles are analyzed in this thesis. Then proposed is a method for creating more robust detection algorithms, specifically an onset detection algorithm suite that splits up the challenges of onset detection into smaller parts and solves them separately.
Issue Date:2017-12
Genre:Other
Type:Text
Language:English
URI:http://hdl.handle.net/2142/97888
Date Available in IDEALS:2017-08-28


This item appears in the following Collection(s)

Item Statistics