ジャーナル論文 Investigating the cochleagram preprocessing and the role of interfered spin-wave-based reservoir computing in speech recognition
Sota Hikasa (author) (この著者で検索)
ORCID https://orcid.org/0009-0005-0888-1783 (unauthenticated)
National Institute for Materials Science
ORCID ;
Wataru Namiki (author) (この著者で検索)
ORCID SAMURAI ;
Daiki Nishioka (author) (この著者で検索)
ORCID SAMURAI ;
Ryo Iguchi (author) (この著者で検索)
ORCID SAMURAI ;
Jiaxuan Chen (author) (この著者で検索)
National Institute for Materials Science
;
Maki Nishimura (author) (この著者で検索)
ORCID SAMURAI ;
Kazuya Terabe (author) (この著者で検索)
ORCID SAMURAI ;
Takashi Tsuchiya (author) (この著者で検索)
ORCID SAMURAI
コレクション

引用
Sota Hikasa, Wataru Namiki, Daiki Nishioka, Ryo Iguchi, Jiaxuan Chen, Maki Nishimura, Kazuya Terabe, Takashi Tsuchiya. Investigating the cochleagram preprocessing and the role of interfered spin-wave-based reservoir computing in speech recognition. Journal of Applied Physics. 2026, 140 (12), 123908. https://doi.org/10.1063/5.0333726

説明:

(abstract)

In this study, we investigated the processing capability required for PRC-only speech recognition by systematically comparing the presence and absence of cochleagram preprocessing and PRC. As the physical reservoir, we employed a nonlinear interfered spin-wave PRC exhibiting input-frequency-dependent responses. Two speech-recognition tasks, spoken digit recognition and speaker classification, were evaluated under all combinations of cochleagram preprocessing and PRC processing. When cochleagram preprocessing was employed, both tasks achieved high-recognition accuracy. In contrast, using only the spin-wave PRC yielded approximately 84.6 % accuracy for speaker classification, whereas spoken digit recognition remained 29.0 %. Furthermore, spoken digit recognition accuracy improved when the speaker was fixed, whereas speaker classification maintained relatively high accuracy even when the spoken digit was fixed. These results indicate that speaker classification and spoken digit recognition require different types of processing in the present spin-wave PRC and provide useful insight into the processing capability required for PRC-only speech recognition.

権利情報:

  • In Copyright

    This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in Sota Hikasa, Wataru Namiki, Daiki Nishioka, Ryo Iguchi, Jiaxuan Chen, Maki Nishimura, Kazuya Terabe, Takashi Tsuchiya; Investigating the cochleagram preprocessing and the role of interfered spin-wave-based reservoir computing in speech recognition. J. Appl. Phys. 28 September 2026; 140 (12): 123908 and may be found at https://doi.org/10.1063/5.0333726.

キーワード: reservoir computing, spin wave, speech recognition

刊行年月日: 2026-09-28

出版者: AIP Publishing

掲載誌:

  • Journal of Applied Physics (ISSN: 00218979) vol. 140 issue. 12 123908

研究助成金:

  • Innovative Science and Technology Initiative for Security JPJ004596
  • Japan Society for the Promotion of Science JP25K17658
  • Ministry of Education, Culture, Sports, Science and Technology JPMXP1225NM5244

原稿種別: 著者最終稿 (Accepted manuscript)

MDR DOI: https://doi.org/10.48505/nims.8665

公開URL: https://doi.org/10.1063/5.0333726

関連資料:

その他の識別子:

連絡先:

更新時刻: 2026-10-19 09:16:28 +0900

MDRでの公開時刻: 2026-10-19 10:30:09 +0900

ファイル名 サイズ
ファイル名 JAP_Hikasa_revised-v2_nohighlight.docx (サムネイル)
application/vnd.openxmlformats-officedocument.wordprocessingml.document
サイズ 23.3MB 詳細