Abstract: Methods, systems, apparatus, and computer-readable media for providing a voice assistant for to leverage information from sensing devices. In some implementations, one or more computers receive measurement data generated by a sensing device associated with a user. The one or more computers store the measurement data in association with an identifier for the user. The one or more computers obtain text of an utterance of the user detected by a client device associated with the user through a voice interface provided by the client device. The one or more computers access the measurement data associated with the user, and in response to obtaining the text of the utterance, generate a response to the utterance based on the text of the utterance and the stored measurement data. The one or more computers provide response data indicating the generated response.