FeelSpeech: A tactile Prosthesis to improve speech perception in persons with hearing loss -- unforeseen problems
The FeelSpeech project aims at creating a novel solution to help hearing-impaired people better understand speech. The envisioned solution is a bracelet providing speech-relevant information through tactile stimulation of the wearer’s wrist. Over time the wearer should develop the ability to incorporate tactile stimulation into the perceptual process, which would result in the subjective experience of hearing better, just like seeing person’s lips in a noisy environment gives an impression of hearing this person’s speech better.
In the past decades a number of attempts have been made to develop “skin-hearing” devices, several of which even showed better speech comprehension improvement than the cochlear implants of that time. However in spite of this progress the skin-hearing devices were not widely adopted. We identified the probable reasons for this lack of adoption, and tried to address them in the FeelSpeech project:
1. The previous devices used vibratory or electrical stimulation of the skin, both of which are known to be unpleasant and irritating especially after prolonged use. Further, the skin is known to habituate and become desensitized after vibration, making vibration an inappropriate way to deliver information.
2. The previous devices used oversimplified processing of the auditory input, which was limited by the state of technology at that time.
3. The previous devices were rather cumbersome, which limited their prolonged use.
4. The previous studies targeted severely hearing impaired individuals, who received nearly no auditory information and thus had difficulties to learn to interpret the tactile stimulation.
In the FeelSpeech project we successfully addressed points 1,2 and 4, and to a certain extent point 3. However, the psychophysical testing we did in the project revealed an additional, unforeseen limitation of skin-hearing devices. This limitation derives from the fact that humans appear to have difficulty integrating tactile and auditory information at a truly perceptual level. Integration rather appears to occur at a slower, more cognitive level, which prevents people from understanding speech at acceptable rates. We concluded that substantial further research in the field is required before this technology will be ready for industrial applications.
