Speech signal processing based on machine learning and complex processors for baby cry detection system

Abstract

Nowadays, a lot of parents has hired a maid to help them to take care of their babies because of the difficulty to take care of newborn babies. Moreover, Parents need to do housework as well as taking careof their newborn babies. However, many maidshave been reported to have some adverse effect on these babies as they grow up. Furthermore, some maids may even expose these babies to too many unexpected risks that would place them at risk. In this research,a baby crying detection system is developed using a Raspberry Pi and Wireless Sensor Network (WSN). Several equipment and controls such as sound sensors, video sensors were integrated and used for baby room surveillance. The Speaking is a communication medium, and speech can be characterized by signals and signals that contain significant information and the information is in sound waveforms. Voice signal is an application of voice signal processing technology. For applications that are in digital form, they rely more on digitally processed speech signals, implement complex technologies, and The framework was programmed using programming language Python 3.6 and Java 8.0, which was used for real-time data transmission and application signaling. Finally, the system will send the data to a remote smartphone. Moreover, the work is integrated with machine learning and IP address to boost the detection mechanism.