Detection of Common Mistakes in Novice Violin Playing
نویسندگان
چکیده
Analyzing and modeling playing mistakes are essential parts of computer-aided education tools in learning musical instruments. In this paper, we present a system for identifying four types of mistakes commonly made by novice violin players. We construct a new dataset comprising of 981 legato notes played by 10 players across different skill levels, and have violin experts annotate all possible mistakes associated with each note by listening to the recordings. Five feature representations are generated from the same feature set with different scales, including two note-level representations and three segmentlevel representations of the onset, sustain and offset, and are tested for automatically identifying playing mistakes. Performance is evaluated under the framework of using the Fisher score for feature selection and the support vector machine for classification. Results show that the Fmeasures using different feature representations can vary up to 20% for two types of playing mistakes. It demonstrates the different sensitivities of each feature representation to different mistakes. Moreover, our results suggest that the standard audio features such as MFCCs are not good enough and more advanced feature design may be needed.
منابع مشابه
MusicJacket: The efficacy of real-time vibro- tactile feedback for learning to play the violin
This research investigates the potential for vibrotactile feedback to enhance motor learning in the context of playing the violin. A prototype has been built which delivers vibrotactile feedback to the arms to indicate to a novice player how to correctly hold the violin and how to bow in a straight manner. This prototype was tested in a pilot user study with four complete beginners. Observation...
متن کاملSeeing the Inaudible. Descriptors for generating objective and reproducible data in real-time for musical instrument playing standard situations
This article describes a method to generate objective and reproducible data to assist instrument teaching and practicing. The method is based on using audio descriptors and their efficient visualisation that assist in the perception of musical parameters difficult to hear. To aid comparison, we defined and recorded a comprehensive database of positive and negative sound examples from the violin...
متن کاملMusical Offset Detection of Pitched Instruments: The Case of Violin
Musical offset detection is an integral part of a music signal processing system that requires complete characterization of note events. However, unlike onset detection, offset detection has seldom been the subject of an in-depth study in the music information retrieval community, possibly because of the ambiguity involved in the determination of offset times in music. This paper presents a pre...
متن کاملDetection of Motor Changes in Violin Playing by EMG Signals
Playing a music instrument relies on the harmonious body movements. Motor sequences are trained to achieve the perfect performances in musicians. Thus, the information from audio signal is not enough to understand the sensorimotor programming in players. Recently, the investigation of muscular activities of players during performance has attracted our interests. In this work, we propose a multi...
متن کاملReal-time Detection and Visualization of Clarinet Bad Sounds
This paper describes an approach on real-time performance 3D visualization in the context of music education. A tool is described that produces sound visualizations during a student performance that are intuitively linked to common mistakes frequently observed in the performances of novice to intermediate students. The paper discusses the case of clarinet students. Nevertheless, the approach is...
متن کامل