#Earballs: Neural transmodal translation https://arxiv.org/abs/2005.13291  "The goal of this work is to propose a mechanism for providing an information preserving mapping that users can learn to use to see (or perceive other information) using their auditory system"

Unlike The vOICe visual-to-auditory sensory substitution, "our system does not seek to encode or preserve the pixel information, but instead the higher-level information extracted by a learned feature embedding"

About The vOICe: "Their audio generation mechanism is based on direct mapping of pixel data from the image to audio frequency, panning, and loudness data (vertical position to frequency, left-right position to scan time (encoded in stereo), and brightness to sound loudness)."

About Earballs: "The GAN successfully found a metric preserving mapping, and in human subject tests, users were able to successfully classify images of faces using only the audio output by our model."