On the neural symphony of speech: decoding speech production with stereo-electroencephalography

Speech is one of the most natural ways for people to share ideas, emotions, and stories. Losing this ability—through a stroke or ALS, for example—can be devastating. This thesis explored the “neural symphony” of speech: how different parts of the brain play together, like instruments in an orchestra, to create sounds, movements, and meaning. By recording brain activity with small electrodes placed deep in the brain, it was shown that speech can be decoded from many different regions. The research also demonstrated how these brain signals can be transformed into sound very fast—an important step toward technology that gives people back their voice. Such systems, called brain-computer interfaces, use a person’s thoughts about speaking to generate artificial speech. The findings in this thesis highlight how understanding the brain’s symphony of speech can both advance neuroscience and help restore communication to those who have lost it.