An HVAC controller may be controlled in response to a natural language audio message that is not recognizable by the HVAC controller as a command, where the natural language audio message is translated into a command recognizable by the HVAC controller. The HVAC controller may be a thermostat and may include a housing that at least partially houses a control module, a microphone, a speaker, a display, and a voice recognition module. The voice recognition module recognizes a predetermined audible trigger from a recorded voice message or streamed voice message, and in response, provide one or more audio clips via the speaker and/or video clips via the display to assist users in operating the thermostat. A user may communicate with the thermostat via the thermostat and/or a remote device.