Restoring speech following total removal of the larynx by a learned transformation from sensor data to acoustics

Created: 2023-10-10T03:32:11-05:00

Return to the Index

This card pertains to a resource available on the internet.

Magnetic sensors are used to read the position of the mouth / larynx features. A neural network is then trained on the sequence of mouth movements to the original sound recorded from the user. The network is then able to synthesize their speech given data out of the implanted sensors.