1. AI infant listener: Deep learning for automatic baby cry recognition and analysis using IoT.
- Author
-
Sinha, Aayush, Dutta, Prasit Kumar, and Jansi, R.
- Subjects
- *
DEEP learning , *PLANT extracts , *ROOT-mean-squares , *ARTIFICIAL intelligence , *INFANTS , *RASPBERRY Pi - Abstract
This paper proposes an IoT-based automatic baby cry detection system using a Raspberry Pi module. The system captures input sound through a condenser microphone, which can be either a baby cry or an environmental sound. The system then extracts features such as Root Mean Square (RMS), Zero Crossing Rate (ZCR), Short-Time Fourier Transform (STFT), Spectral Centroid (SC), and Mel-Frequency Cepstral Coefficients (MFCC) from the sound data. These features are classified using the Keras Sequential API. If the output from the classification module is detected as a baby cry sound, the system sends an alert to the parent/caretaker using the Telebot application. The proposed system can be useful for busy parents/caretakers who may not always be able to attend to their child's needs. The results show that the proposed system achieved high accuracy in distinguishing baby cries from environmental sounds, demonstrating its potential for real-world applications. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF