Disclosed herein are methods, systems, and devices for discontinuously recording audio input representative of an audio environment. The audio environment can be an environment of a recipient of a hearing prosthesis and the discontinuously recorded audio input can be used to characterize the audio environment of the recipient, to control settings of the hearing prosthesis, or to provide some other application. Discontinuously recording the audio input can provide a recording that characterizes the audio environment of the recipient while protecting the privacy of individuals speaking nearby the recipient. Discontinuously recording the audio input may include recording discontinuous time segments of the audio input such that the semantic content of speech within the audio input, the identity of speakers in the audio input, or other private information present in the audio input is masked.