How can speech centres in the brain be used to control a synthetic voice?

Researchers have developed advanced devices that generate a synthetic voice by interfacing with the speech centres of the brain. This process relies on the following methodology:

• Mapping and Correlation: Scientists map the sounds produced by the voices of healthy individuals to the specific movements of the vocal tract required to create those sounds. Using this data, they develop algorithms that associate the brain activity generated during speech with those specific vocal tract movements.1
• Neural Decoding: By monitoring neuronal activity in the brain, these devices can interpret the signals intended for speech. Even when a person silently mimes words, the system decodes the brain's activity into the corresponding vocal tract movements.1
• Synthetic Generation: Once the brain signals are correlated with vocal tract movements, the device uses that information to produce a synthetic version of the individual's voice. In early trials, this technology has successfully generated understandable sentences from silent speech.1

This approach leverages the brain's ability to control complex motor acts, such as those coordinated by the motor cortex and Broca area, to bypass physical limitations and restore communication.123