Emerging technologies, such as the Internet of Things (IoT) and Artificial Intelligence (AI), are transforming industrial processes, fostering well-being, and ensuring occupational safety in paradigms such as Industry 4.0 and Society 5.0. In noisy industrial environments, prolonged exposure to intense noise from heavy machinery poses significant auditory risks, highlighting the need for effective hearing protection. The integration of IoT technologies into personal protective equipment (PPE) systems is becoming increasingly crucial, as it enables real-time monitoring and adaptive responses to environmental changes. The system features AI-driven adaptive noise-filtering headphones that enhance speech intelligibility. Integrated with a software framework, it delivers synthesized audio alerts to improve safety and situational awareness, dynamically adjusting to real-time noise levels. This work focuses on adapting an existing Voice Activity Detection (VAD) model, Silero VAD, to the system, in order to ensure real-time performance and seamless integration. The system includes a dashboard that allows users to evaluate the noise filtering in action. Serving as a functional demonstrator, it also assesses the interaction of Silero VAD with the DSP filter. Through extensive evaluations under various acoustic conditions, we analyzed the effectiveness of this integration in industrial noise environments to optimize system performance.
Smart Hearing Protection System based on IoT and AI for Industrial Noisy Environments
Davide RolloPrimo
;Mattia CotardoSecondo
;Ilaria Sergi;Teodoro Montanaro;Hossem Eddine Hafidi;Luigi Patrono
Ultimo
2025-01-01
Abstract
Emerging technologies, such as the Internet of Things (IoT) and Artificial Intelligence (AI), are transforming industrial processes, fostering well-being, and ensuring occupational safety in paradigms such as Industry 4.0 and Society 5.0. In noisy industrial environments, prolonged exposure to intense noise from heavy machinery poses significant auditory risks, highlighting the need for effective hearing protection. The integration of IoT technologies into personal protective equipment (PPE) systems is becoming increasingly crucial, as it enables real-time monitoring and adaptive responses to environmental changes. The system features AI-driven adaptive noise-filtering headphones that enhance speech intelligibility. Integrated with a software framework, it delivers synthesized audio alerts to improve safety and situational awareness, dynamically adjusting to real-time noise levels. This work focuses on adapting an existing Voice Activity Detection (VAD) model, Silero VAD, to the system, in order to ensure real-time performance and seamless integration. The system includes a dashboard that allows users to evaluate the noise filtering in action. Serving as a functional demonstrator, it also assesses the interaction of Silero VAD with the DSP filter. Through extensive evaluations under various acoustic conditions, we analyzed the effectiveness of this integration in industrial noise environments to optimize system performance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


