Playing Virtual Musical Drums by MEMS 3D Accelerometer Sensor Data and Machine Learning


Shaikh Farhad Hossain, Kazuhisa Hirose, Shigehiko Kanaya and Md. Altaf-Ul-Amin, Nara Institute of Science and Technology (NAIST), Japan


In our life, music is a vital entertainment part whose important elements are musical instruments. Forexample, the acoustic drum plays a vital role when a song is sung. With the modern era, the style of themusical instruments is changing by keeping identical tune such as an electronic drum. In this work, wehave developed "Virtual Musical Drums" by the combination of MEMS 3D accelerometer sensor data and machine learning. Machine learning is spreading in all arena of AI for problem-solving and the MEMS sensor is converting the large physical system to a smaller system. In this work, we have designed eight virtual drums for two sensors. We have found a 91.42% detection accuracy within the simulation environment and an 88.20% detection accuracy within the real-time environment with 20% windows overlapping. Although system detection accuracy satisfaction but the virtual drum sound was nonrealistic. Therefore, we implemented a 'multiple hit detection within a fixed interval, sound intensity calibration and sound tune parallel processing' and select 'virtual musical drums sound files' based on acoustic drum sound pattern and duration. Finally, we completed our "Playing Virtual Musical Drums" and played the virtual drum successfully like an acoustic drum. This work has shown a different application of MEMS sensor and machine learning. It shows a different application of data, sensor and machine learning as music entertainment with high accuracy.


Virtual musical drum, MEMS, SHIMMER, support vector machines (SVM) and k-Nearest Neighbors (kNN)

Full Text  Volume 10, Number 3