The Methods of Realizing Baby Crying Recognition and Intelligent Monitoring Based on DNN-GMM-HMM

I118 this paper, we propose a crying recognition model based on Deep Neural Network Gaussian Mixture Model Hidden Markov Model (DNN-GMM-HMM), using the acoustical model established by GMM-HMM to decode the reduced dimensionality features, and with the help of DSP Technology, the meaning of baby crying can be accurately analyzed. To attain the goal of intelligent monitoring of babies, this paper mainly adopts STM32, DSP technology and Bluetooth 4.0BLE technology to realize the function of baby crying identification, heart rate monitoring, and security alerts.