AES E-Library

Impression Evaluation Model for Button Sounds Using a Neural Network

This paper presents an impression evaluation model for button sounds generated when users press the buttons on car audio equipment using a neural network. The dynamic characteristics of 11 kinds of button sounds obtained by their wavelet transform frequencies and sound pressure values are fed into the network model inputs. The model then responds with three factor scores, “esthetic”, “force” and “metallic”, and an evaluation value of “offensive - pleasant” as the outputs. By analyzing the inside functions of the neural network after training, we confirmed the model acquired a mechanism that extracted four impression evaluation values from the sound characteristics, thus showing the model could attain automation of button sound design.

Authors: Onishi, Gen; Ishimitsu, Shunsuke; Sakamoto, Koji; Arai, Takayuki; Yoshimi, Toshikazu; Fujimoto, Yuichi; Kawasaki, Kenichi
Affiliations: Hiroshima International University; Hiroshima City University; University of Hyogo; Pioneer Corporation(See document for exact affiliation information.)
AES Conference:34th International Conference: New Trends in Audio for Mobile and Handheld Devices (August 2008)
Paper Number:1 Import into BibTeX
Subject:Audio for Mobile & Handheld Devices: Testing and Evaluation

Click to purchase paper or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!

This paper costs $20 for non-members, $5 for AES members and is free for E-Library subscribers.

Learn more about the AES E-Library

E-Library Location: (CD 34thPapers)   /34/aes34-000006.pdf

Start a discussion about this paper!