Poster: MobiEar-Building an Environment-independent Acoustic Sensing Platform for the Deaf using Deep Learning

ABSTRACT

Acoustic 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].

References

1. N. D. Lane, P. Georgiev, and L. Qendro. Deepear: robust smartphone audio sensing in unconstrained acoustic environments using deep learning. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing.

Index Terms

• Hardware

  • Communication hardware, interfaces and storage

    • Sound-based input / output

• Human-centered computing

  • Human computer interaction (HCI)

  • Ubiquitous and mobile computing

    • Ubiquitous and mobile computing theory, concepts and paradigms

      • Mobile computing