Approaches to enhance the performance of FeFET reservoir computing with parallel data processing for speech recognition

Eishin Nako; Kasidit Toprasertpong; Ryosho Nakane; Mitsuru Takenaka; Shinichi Takagi

Presentation information

Oral presentation

FS Focused Session "AI Electronics" » FS.1 Focused Session "AI Electronics"

[23p-M206-1~13] FS.1 Focused Session "AI Electronics"

Fri. Sep 23, 2022 1:30 PM - 5:00 PM M206 (Multimedia Research Hall)

Shinichi Takagi(Univ. of Tokyo), Norio Sato(NTT), Yoshihiko Horio(東北大)

4:30 PM - 4:45 PM

[23p-M206-12] Approaches to enhance the performance of FeFET reservoir computing with parallel data processing for speech recognition

〇Eishin Nako¹, Kasidit Toprasertpong¹, Ryosho Nakane¹, Mitsuru Takenaka¹, Shinichi Takagi¹ (1.Univ. Tokyo)

Keywords:reservoir computing, FeFET

We have proposed a novel scheme of FeFET reservoir computing with parallel data processing for speech recognition and we have achieved the classification accuracy of 82.6% in spoken digit recognition. In this study, we consider approaches to enhance the performance and improve the classification accuracy by combining approaches.