Mobile intelligent terminal speaker identification for real-time monitoring system of sports training

摘要: Mobile intelligent terminal speaker identification for real-time monitoring system of sports training is presented in this manuscript. The purpose recognition to use speaker’s voice characteristics identify the identity. No two people world have exactly same characteristics, because from perspective human physiological structure, different people’s vocal shapes, throat sizes, and other factors related pronunciation are different, which makes it possible recognize speech by speech. From technology level, proposed framework contains four core aspects, namely: identification, motion gesture detection, heart rate detection breath detection. In information collection part, we adopt ZigBee + WIFI wireless transmission, GPRS data transmission Bluetooth communication joint capture accurate rrealize dual-network dual-pass function, TD-SCDMA receiver must also work at time when GSM radio module working. propose novel deep neural network revise architecture. experiment compared with state-of-the-art models proven robustness framework.