Speech-to-touch sensory substitution: a 10-decibel improvement in speech-in-noise understanding after a short training

Abstract

Understanding speech in background noise is challenging. Wearing face-masks during COVID19-pandemics made it even harder. We developed a multi-sensory setup, including a sensory substitution device (SSD) that can deliver speech simultaneously through audition and as vibrations on fingertips. After a short training session, participants significantly improved (16 out of 17) in speech-in-noise understanding, when added vibrations corresponded to low-frequencies extracted from the sentence. The level of understanding was maintained after training, when the loudness of the background noise doubled (mean group improvement of ~ 10 decibels). This result indicates that our solution can be very useful for the hearing-impaired patients. Even more interestingly, the improvement was transferred to a post-training situation when the touch input was removed, showing that we can apply the setup for auditory rehabilitation in cochlear implant-users. Future wearable implementations of our SSD can also be used in real-life situations, when talking on the phone or learning a foreign language. We discuss the basic science implications of our findings, such as we show that even in adulthood a new pairing can be established between a neuronal computation (speech processing) and an atypical sensory modality (tactile). Speech is indeed a multisensory signal, but learned from birth in an audio-visual context. Interestingly, adding lip reading cues to speech in noise provides benefit of the same or lower magnitude as we report here for adding touch.