Automatic Assessment of Tone Quality in Violin Music Performance

Abstract

The automatic assessment of music performance has become an area of increasing interest due to the growing number of technology-enhanced music learning systems. In most of these systems, the assessment of musical performance is based on pitch and onset accuracy, but very few pay attention to other important aspects of performance, such as sound quality or timbre. This is particularly true in violin education, where the quality of timbre plays a significant role in the assessment of musical performances. However, obtaining quantifiable criteria for the assessment of timbre quality is challenging, as it relies on consensus among the subjective interpretations of experts. We present an approach to assess the quality of timbre in violin performances using machine learning techniques. We collected audio recordings of several tone qualities and performed perceptual tests to find correlations among different timbre dimensions. We processed the audio recordings to extract acoustic features for training tone-quality models. Correlations among the extracted features were analyzed and feature information for discriminating different timbre qualities were investigated. A real-time feedback system designed for pedagogical use was implemented in which users can train their own timbre models to assess and receive feedback on their performances.

Technology Use and Attitudes in Music Learning

“…The low scores for the skills of instrument handling, maintaining good posture, and avoiding injury all speak to the physical aspects of technology, an area that could see further growth soon due to the increasing development of optical and wearable sensors for music performance and corresponding experimental pedagogical applications (e.g., Ng et al, 2007;Van der Linden et al, 2009;Johnson et al, 2010;Volpe et al, 2017). The low score for the skill of good tone or timbre may also speak to the complexity of the construct and a lack of marketready technologies to analyze and develop this skill, although new strides are being made in this area (Himonides, 2009;Giraldo et al, 2017Giraldo et al, , 2019.…”

Evaluación psicoacústica y profesional sobre la interpretación vocal en estudiantes de canto

“…La evaluación automática de la calidad del tono en la interpretación musical del violín se ha convertido en un tópico de creciente interés, ya que la obtención de criterios cuantificables para la evaluación de la calidad en la emisión sonora constituye, sin duda, un reto. Giraldo et al (2019) implementan tecnología diseñada para uso pedagógico en la que los usuarios pueden entrenar la emisión de su propio timbre y recibir retroalimentación de sus interpretaciones. Volpe et al (2017) han creado un corpus multimodal que recoge, a partir de la grabación audio-visual de interpretaciones musicales realizadas por expertos, datos cinemáticos (velocidad, aceleración, sobre-aceleración, curvatura de trayectorias relevantes de manos y cabeza, distancias entre huesos, energía cinética, entre otros) y características de nivel superior (ligereza, balanceo, tensión, brusquedad o coordinación), para su cuantificación y análisis en profundidad.…”

SonoSpace

“…Several systems especially focus on characterizing musical performance aspects beyond pitch and onset, such as timbre and articulation [5,8,19]. For example, Giraldo et al [5] proposed a system that automatically assesses the quality of timbre in violin sounds using machine-learning techniques. Knight et al [8] used support vector machines to assess the trumpet tone quality.…”