As a model of a “natural” human-machine interface, we present a flexible hybrid electronic (FHE) large-area/integrated circuit system which can separate the separate the speech of individuals when multiple people are speaking at once and sense human gestures such as hand swipes from a distance on the order of 50 cm. The system built as a single “roll-up” sensing sheet for easy deployment, with a size on the scale of meters. The voice separation relies on a linear array of flexible microphones feeding into a modified “inverse beam-forming” algorithm. A “blind” option of the algorithm does not require training or knowledge of the speaker locations. The gesture sensing relies on ultra-sensitive capacitive sensing with flexible electrode sensors. Minute capacitance changes can be read out by small changes in the frequency of an LC oscillator. The user does not use “artificial” aids such as clickers or lapel microphones, making it an intuitive and natural approach.