A Preliminary Study on Speaker Identification Using Business Card-Type Sensors
Shunpei Yamaguchi, Ritsuko Oshima, Jun Oshima, Takuya Fujihashi, Shunsuke Saruwatari, Takashi Watanabe · 2021
Speaker identification using mobile devices requires to sample a microphone signal at a high sampling rate. High hardware cost is inevitable to process data at a high sampling rate. In this paper, we propose a novel business card-type sensor and a speaker identification algorithm to accurately identify speakers. The algorithm identifies speakers by analyzing the sound pressure data sampled at 100 Hz by the sensors. The proposed method identifies speakers with 100 % accuracy using the users' sound pressure data.