JERUSALEM -- Israeli researchers developed an artificial intelligence (AI) method to enable paralyzed, silent patients to "speak" only through the power of thought, Tel Aviv University (TAU) said in a statement on Tuesday.

In a study conducted by TAU and Tel Aviv Sourasky Medical Center, published in the journal Neurosurgery, depth electrodes implanted in a participant's brain transmitted electrical signals to a computer, which then vocalized two syllables the participant was asked to imagine.

The researchers noted that this marks the first time that parts of speech have been linked to the activity of individual cells in the brain regions from which they were recorded.

TAU stated that these findings offer hope for enabling individuals who are completely paralyzed due to ALS, brainstem stroke, or brain injury to express themselves through artificial speech.

This advancement may also allow, for example, training a computer for an ALS patient in the early stages of the disease, while they are still able to speak, enabling interpretation after the patient loses muscle control.

Initially, the researchers asked the participant, an epilepsy patient with electrodes already implanted in his brain before surgery, to say the syllables "a" and "e," recording the brain activity as he articulated these sounds.

Using deep learning and machine learning techniques, the researchers trained AI models to identify specific brain cells whose electrical activity indicated the intention to say "a" or "e."

Once the computer learned to recognize the patterns of electrical activity associated with these two syllables in the patient's brain, he was instructed to merely imagine saying the two syllables.

Subsequently, the computer translated the electrical signals and played pre-recorded sounds of "a" or "e" accordingly.

The researchers said that their next ambition is to get to complete speech, but even two different syllables can enable a fully paralyzed person to signal "yes" and "no."

They concluded that this marks a significant step towards developing a brain-computer interface capable of replacing the brain's pathways for speech production.