Real-Time Hospital Noise Detection and Classification Using IoT Devices, MFCC, and Artificial Neural Networks

Hospital noise is a critical issue affecting patient well-being and healthcare outcomes. In this paper, we propose an innovative system for hospital noise detection and segregation leveraging Internet of Things (IoT) technology, Mel-frequency cepstral coefficients (MFCC), and Artificial Neural Networks (ANN). Our approach involves deploying IoT devices equipped with microphones in hospital environments to continuously capture audio data. Pre-processing techniques are applied to the collected audio sam ples to remove noise and normalize signals. MFCC features are then extracted from the pre-processed data to capture unique sound characteristics. These features serve as inputs to a supervised ANN model trained to classify noise into hospital and non-hospital categories. The performance of the model is evalu ated using a validation dataset, and upon achieving satisfactory results, the model is deployed in the IoT infrastructure for real-time noise analysis. Con tinuous monitor- ing and periodic fine-tuning ensure adaptability to evolving noise patterns.