According to United Nations’ World Health Organization, there are approximately 278 million people worldwide suffering from moderate to profound hearing loss. It is not surprising that many of those people have particular difficulty with telephone communications and programs through the Internet. When telephone conversations are conducted via computer networks using the Internet Protocol, ambient noise and acoustic echoes often impede the conversation. For the hearing impaired, it is especially problematic - most of the time they need to increase the volume to try and follow the conversation. However, by doing so, the background noises are also intensified and signal frequencies become virtually intolerable. In response to this growing problem, developers at the Fraunhofer Institute for Digital Media Technology IDMT in Oldenburg have come up with a digital solution. The Speech-Improved Telephony project utilizes algorithms typically used for hearing aids to detect background noise and reduce it to a minimum. The system is adjusted to the individual user, as each hearing impaired person has quite specific frequencies that are difficult for them to hear. Since the system reduces background noise, it could also be beneficial to businesses making calls from an open plan office or for busy call centers.
The system can be individually programmed for each call, so that it delivers a consistently audible sound. “One particular challenge is to figure out how users can moderate the algorithms themselves in a user-friendly manner," explained Stefan Goetze of the Hearing, Speech and Audio Technology project group at IDMT. "For seniors in particular, simple methods for making adjustments needed to be found. We were able to solve this on a test telephone through a special display. Two audio signals with different sound were visualized through flowers. By pressing on the flowers, the seniors can regulate the desired sound. This automatically adjusts the algorithm parameters to the hearing ability of the individual user."
Fraunhofer's system can be applied to most audio devices and scientists have already installed them on an iPod Touch, a telephone system, a video conferencing system and a television. The public, however, are not likely to be able to get their hands on one for at least another two years.
The researchers will display a video conferencing system in which their algorithms are installed at the IFA consumer electronics show in Berlin this September.
