Rhythm-tracking for guitar players using smart watch device.

Bc. Artemiy Mikheev

Master's thesis

Abstract:

The thesis is dedicated to research that may enable development of a rhythm-tracker for guitar players. As the possible application is intended to be delivered onto wearable platforms, Android Watch device was used during the research. For this intent, sensor data were collected from the smart watch sensors (namely 6-axis MEMS, magnetometer and microphone) during the specifically designed set of experiments, targeted at covering as much of potential inputs variation as possible. The experimental data were collected with the help of physical equipment set, supported by auxiliary software, e.g. SensorCap application. Initial set of experiments proved to result in datasets which were insufficiently exact to be interpreted, therefore additional validation tool in form of synchronized slow-motion camera footage was introduced to align the data timestamps with actual physical events properly. Additionally, certain types of events have been discovered to produce unintelligible data due to high level of noise captured by smart watch sensors. Second set of experiments, supported by validation timestamps, was processed through the developed pipeline with minor variations, later the most suitable criterion was detected among the ones suggested by the sensor arsenal and the variety of derived values. The correlation has been discovered between validation set pulses and MEMS data from two specific axes, so this may be a foundation for the actual end-user-friendly solution.

Keywords

smartwatch, rhythm, IoT, sensor, data analysis, application, algorithm, music, tracker, tracking, guitar

Language used: English

Date on which the thesis was submitted / produced: 6. 4. 2020

Thesis defence

• Supervisor: Ing. Jan Masner, Ph.D.

Czech University of Life Sciences Prague

Faculty of Economics and Management

Master programme / field:
Systems Engineering and Informatics / Informatics