Loudspeaker Sound Quality Evaluation Using EEG and GMDH-type Neural Network

カテゴリ: 部門大会

論文No: OS4-4

グループ名: 【C】平成25年電気学会電子・情報・システム部門大会講演論文集

発行日: 2013/09/04

タイトル(英語): Loudspeaker Sound Quality Evaluation Using EEG and GMDH-type Neural Network

著者名: Kiminobu Nishimura(Keio University),Yasue Mitsukura(Keio University)

著者名(英語): Kiminobu Nishimura(Keio University),Yasue Mitsukura(Keio University)

キーワード: EEG|Neural network|Group method of data handling|Sound quality evaluation

要約(日本語): In this paper, we propose a sound quality evaluation system using electroencephalogram (EEG) and group method of data handling (GMDH) type neural network. Recently, EEG is used in various applications to evaluate human preferences, and we focus on sound quality evaluation. We prepared EEG samples for the GMDH-type neural network to recognise 3 typical types of sound using acoustic equalizers which were used to create the training data. The results showed that the method using GMDH-type neural network improved recognition rate compared to conventional method which uses.

PDFファイルサイズ: 12,264 Kバイト