Learning violin is a challenging task requiring execution of pitch tasks with the left hand using a strong aural feedback loop for correctly adjusting pitch, concurrent with the right hand moving a bow precisely with correct pressure across strings. Real-time technological assistance can help a student gain feedback and understanding helpful for learning and maintaining motivation. This thesis presents real-time low-cost low-latency violin augmentations that can be used to assist learning the violin along with other real-time performance tasks. To capture bow performance, we demonstrate a new means of bow tracking by measuring bow hair de ection from the bow hair being pressed against the string. Using near- eld optical sensors placed along the bow we are able to estimate bow position and pressure through linear regression from training samples. For left hand pitch tracking, we introduce low cost means for tracking nger position and illustrate the combination of sensed results with audio processing to achieve high accuracy low-latency pitch tracking. We subsequently verify our new tracking methods' e ectiveness and usefulness demonstrating low-latency note onset detection and control of real-time performance visuals. To help tackle the challenge of intonation, we used our pitch estimation to develop low latency pitch correction. Using expert performers, we veri ed that fully correcting pitch is not only disconcerting but breaks a violinist's learned pitch feedback loop resulting in worse asplayed performance. However, partial pitch correction, though also linked to worse as-played performance, did not lead to a signi cantly negative experience con rming its potential for use to temporarily reduce barriers to success. Subsequently, in a study with beginners, we veri ed that when the pitch feedback loop is underdeveloped, automatic pitch correction did not signi cantly hinder performance, but o ered an enjoyable low-pitch error experience and that providing an automatic target guide pitch was helpful in correcting performed pitch error.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:765983 |
| Date | January 2017 |
| Creators | Pardue, Laurel S. |
| Publisher | Queen Mary, University of London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://qmro.qmul.ac.uk/xmlui/handle/123456789/25934 |
Page generated in 0.0022 seconds