A Human Absolute Pitch Model for Identifying Musical Pitch

Takeshi Shono, Takahiro Emoto, Udantha R. Abeyratne, Masatake Akutagawa, and Yohsuke Kinouchi

Keywords

Absolute Pitch, Auditory image model, Neural network

Abstract

Absolute pitch (AP) is the ability to identify or produce a pitch of a sound without a reference sound. In this study, we propose a novel method to identify musical pitch using an absolute pitch model, which models chroma perception of AP possessors. The proposed method is based on the use of an auditory image model (AIM) and neural network (NN). The performance of the proposed model was evaluated by using the musical instrumental sound database of musical notes for each instrument. We show that the proposed method can contribute to identifying growl style sounds, which degrade the accuracy of pitch estimation. We believe that the proposed method will improve the accuracy of biological sound coding and may be useful for the biological-sound- based diagnosis.

Important Links:



Go Back


IASTED
Rotating Call For Paper Image