The long-term goal of this project is the development of an adaptable digital-signal-processing-based alerting system for person with hearing disabilities that will be used to make the owner aware of important sounds (e.g.. doorbell, telephone, smoke detector, infant crying) in his/her hone, office car classroom, or other environment. The product must be capable of detecting these sounds in """"""""acoustically hostile"""""""" environments such as when a television is operating. The system will be flexible and portable and will be trained by the owner to recognize an array of sounds of the owner's choosing. The critical technological issues to be explored are the development of a robust signal detection system based on artificial neural network leaning algorithms to """"""""teach"""""""" the system new sounds (and to extinguish alerts to inappropriate ones), and, ultimately, the design and manufacture of application specific integrated circuitry to implement the algorithms.
This portable alerting system with the capability to """"""""learn"""""""" sounds and adapt to various environments will not only improve the lives of deaf and hearing impaired people in the home environment,but will also enhance employment opportunities for and improve access to educational opportunities for people with hearing disabilities. These features have distinct commercial advantages over all currently available and will have broad appeal among the 28 million deaf and profoundly hearing disabled individuals U.S. alone.
