Recently, Jon Hamilton of NPR’s All Things Considered interviewed Dr. Edward Chang, one of the neurosurgeons and investigators involved in a study focused on decoding cortical activity into spoken words.
Currently, those who cannot produce speech rely upon technology that allows them to use eye gaze to produce synthesized speech one letter at a time. While this gives those who otherwise could not speak a voice, it is considerably slower than natural speech production.
In the current study, cortical electrodes gathered information as subjects read hundreds of sentences. The electrodes monitored various portions of the cortex involved in speech production. This information was processed and resulted in intelligible synthesized speech.
Reference
Anumanchipalli GK, Chartier J, Change E. (2019) Speech synthesis from neural decoding of spoken sentences. Nature568:493–498.
Recent Posts
No Surprises Act Reform: One to Watch in the Lame-Duck Session
As Congress prepares for the lame-duck session following the November midterm elections, health care legislation could be among the issues lawmakers seek to address before…
In Memoriam: Jerry Northern, PhD
Dr. Jerry Northern made incomparable contributions to his profession and to the American Academy of Audiology throughout his career. His recent passing is a loss…
AMA RUC Upcoming Surveys
Over the next several weeks, a significant number of CPT® codes used to report audiology function tests will undergo review through the AMA/Specialty Society Relative…



