Abstract: Techniques, including a system, device and method, for decoding and using silent speech signals are provided. The techniques may record silent speech signals using a wearable device, worn by a user. The techniques may analyze the silent speech signals to determine one or more outputs including words or phrases from the silent speech signals. The techniques may comprise using machine learning models to determine the one or more outputs. The techniques may involve performing one or more actions in response to the one or more outputs. The one or more actions may include natural language processing, executing a command on a device and providing information, among other actions. The techniques may involve generating an audio signal for communication from silent speech signals.

Abstract: A headgear for magnetoencephalography includes a body defining a plurality of ports, where the body includes a first portion and a second portion; an adjustment mechanism coupled to the first portion and the second portion of the body and configured to adjust a separation between the first and second portion to facilitate fitting the headgear to a head of a user; and a plurality of optically pumped magnetometer (OPM) modules, where each of the OPM modules includes at least one vapor cell and is configured to be removably inserted into a one of the ports of the body, where each of the OPM modules is configured for coupling to a light source for receiving light.