Downloads: 0
Research Paper | Computer Science | Volume 15 Issue 4, April 2026 | Pages: 1183 - 1186 | India
Real-Time Emergency Alert System for Deaf People Using AI
Abstract: Deaf and hard-of-hearing individuals face significant challenges in detecting critical environmental sounds such as alarms, sirens, and doorbells, which are essential for personal safety. This paper presents an AI-based real-time emergency alert system that identifies important sounds and provides immediate notifications through vibration and visual alerts. The system is developed using Flutter for cross-platform mobile application development and TensorFlow Lite for efficient on-device sound classification. Audio features such as MFCC and spectrograms are extracted using Python and Librosa to train the model. The system operates fully offline, ensuring privacy, low latency, and reliability. It enhances safety, independence, and accessibility for hearing-impaired individuals and demonstrates the real-world application of artificial intelligence in assistive technology.
Keywords: Deaf and hard-of-hearing individuals face significant challenges in detecting critical environmental sounds such as alarms, sirens, and doorbells, which are essential for personal safety. This paper presents an AI-based real-time emergency alert system that identifies important sounds and provides immediate notifications through vibration and visual alerts. The system is developed using Flutter for cross-platform mobile application development and TensorFlow Lite for efficient on-device sound classification. Audio features such as MFCC and spectrograms are extracted using Python and Librosa to train the model. The system operates fully offline, ensuring privacy, low latency, and reliability. It enhances safety, independence, and accessibility for hearing-impaired individuals and demonstrates the real-world application of artificial intelligence in assistive technology.
How to Cite?: Aswin P Sunil, Preethi Thomas, "Real-Time Emergency Alert System for Deaf People Using AI", Volume 15 Issue 4, April 2026, International Journal of Science and Research (IJSR), Pages: 1183-1186, https://www.ijsr.net/getabstract.php?paperid=SR26419223453, DOI: https://dx.dx.doi.org/10.21275/SR26419223453