Measurement, Recognition, and Visualization of Piano Pedaling Gestures and Techniques

Liang B, Fazekas G, Sandler M. Measurement, Recognition, and Visualization of Piano Pedaling Gestures and Techniques. J Audio Eng Soc. 2018;66(6):448-456. doi:10.17743/jaes.2018.0035

@article{Liang2018_19584,
  author = {Liang, Beici and Fazekas, György and Sandler, Mark},
  title = {{Measurement, Recognition, and Visualization of Piano Pedaling Gestures and Techniques}},
  journal = {Journal of the Audio Engineering Society},
  volume = {66},
  number = {6},
  pages = {448--456},
  year = {2018},
  month = jun,
  publisher = {Audio Engineering Society},
  doi = {10.17743/jaes.2018.0035},
  url = {https://doi.org/10.17743/jaes.2018.0035}
}

TY  - JOUR
TI  - Measurement, Recognition, and Visualization of Piano Pedaling Gestures and Techniques
AU  - Liang, Beici
AU  - Fazekas, György
AU  - Sandler, Mark
T2  - Journal of the Audio Engineering Society
J2  - J. Audio Eng. Soc.
VL  - 66
IS  - 6
SP  - 448
EP  - 456
PY  - 2018
DA  - 2018/06/06
DO  - 10.17743/jaes.2018.0035
UR  - https://doi.org/10.17743/jaes.2018.0035
PB  - Audio Engineering Society
LA  - en
AB  - When playing the piano, pedaling is one of the important techniques that lead to expressive performance, comprising not only the onset and offset information that composers often indicate in the score, but also gestures related to the musical interpretation by performers. This research examines pedaling gestures and techniques on the sustain pedal from the perspective of measurement, recognition, and visualization. Pedaling gestures can be captured by a dedicated measurement system where the sensor data is simultaneously recorded alongside the piano sound under normal playing conditions. Recognition is comprised of two separate tasks on the sensor data: pedal onset/offset detection and classification by technique. The onset and offset times of each pedaling technique were computed using signal processing algorithms. Based on features extracted from every segment when the pedal is pressed, the task of classifying the segments by pedaling technique was undertaken using machine-learning methods. High accuracy was obtained by cross validation. The recognition results can be represented using novel pedaling notations and visualized in an audio-based score-following application.
ER  -