STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Mechatronics and Automation

Automatic Baby Cry Detector with sleep music player (ABCD)

  1. Kanha Malviya
  2. Navneet Kaur
  3. Aniket Brahamne
  4. Nikhil Kumar
  5. Mohit Suryawanshi

Abstract

In today’s world, our lives have become moredependent on technology. One problem which caught our eyesis when parents have to leave their wards (aged between 3months to 2 years) alone due to some essential tasks for a shortduration, the baby goes unmonitored. Normally, parents dothis by leaving their child asleep. In many cases, the childwakes up and starts to cry. In absence of loved ones, babiesneed immediate calmness relief. A baby’s cry can becharacterized by its natural periodic tone and the change of itsvoice. This study of sound recognition involves featureextraction and classification by determining the sound pattern.The solution to this problem is, Automatic Baby Cry Detectorwith sleep music player (ABCD), which is programmed onPython and the heart of this device is an Audio learning model.Of this model, a Deep Learning algorithm works by firstextracting MFCCs of the filtered audio. Then, these soundfeatures (extracted by MFCCs) are predicted by the trainedModel. This trained model identifies the Baby Cry byclassifying audio using K-Nearest Neighbor (k-NN) method.Based on the probability confirmer (for model accuracy), thismodel confirms a baby’s cry. Then the program can furtherplay a lullaby (programmable) and also alert the parents bysending a notification on their phone.

Keywords

Support