Extended Data Fig. 1: Median original and decoded spectrograms.
From: Speech synthesis from neural decoding of spoken sentences
a, b, Median spectrograms, time-locked to the acoustic onset of phonemes from original (a) and decoded (b) audio (/i/, n = 112; /z/, n = 115; /p/, n = 69, /ae/, n = 86). These phonemes represent the diversity of spectral features. Original and decoded median phoneme spectrograms were well-correlated (Pearson’s r > 0.9 for all phonemes, P = 1 × 10−18).
