Ruize Yu¹ and Yu Sun², ¹USA, ²California State Polytechnic University, USA

Due to the ever growing popularity of music as a part of everyday life, and with the continuous advances in AI technology, it is now possible for computers to listen to and recognize music [1]. However, there still exist limitations on machines’ ability to recognize audio. This paper proposes an application to simplify the process of music transcription and reduce its runtime [2]. This application was tested in a different range of settings and evaluated. The results show what can be further improved on this application.

MIDI, Pitch Recognition, Magenta.