Abstract: A speech-processing system receives input data representing text. One or more encoders trained to predict audio properties corresponding to the text process the text to predict those properties. A speech decoder processes phoneme embeddings as well as the predicted properties to create data representing synthesized speech.

Abstract: A device for modulating the phase of a light beam includes a matrix of elementary cells, called pixels, coupled to a circuit for addressing the pixels, the device further comprising a set of so-called lateral electrodes extending in a so-called vertical direction (Y) at right angles to the alignment direction (Xa) and configured to apply, for each pixel and via at least two lateral electrodes, a so-called acceleration voltage generating a lateral electrical field (Era) substantially parallel to the alignment direction, in a vector allowing an accelerated return of the liquid crystal molecules to their orientation of rest, the acceleration voltage being configured to be applied in a phase called acceleration relaxation phase, when the activation voltage is no longer applied.