Poster: Mobiear-Building An Environment-Independent Acoustic Sensing Platform For The Deaf Using Deep Learning

Sicong Liu,Junzhao Du

DOI: https://doi.org/10.1145/2938559.2948831

2016-01-01

Abstract:ABSTRACTAcoustic alarms have been credited with saving thousands of lives from fires, gas leakage and electric leakage each year. By broadcasting sound with different tones, loudness and timbres, acoustic alarms keep people aware of surroundings, inform them of serendipitous events, and notify them critical information. However, maintaining the safety awareness through the acoustic alarm is difficult for people who are deaf or less sensitive to acoustic signals. They are too often among the last to access important information even when they are in dangers, especially when they stay alone. By leveraging the microphone on commodity smartphones, universal sound awareness applications are becoming possible. Deep learning models have large leaps in accuracy and robustness[1].