A method for automatic audio volume control on a robot is presented. The robot can deliver its audio output at a comfortable and intelligible level to the user according to user's distance and background noise intensity in user's environment. The user's distance is estimated, by using a camera with known focal length and resolution, by using a stereo camera with known focal length and distance between lenses, or by using an electronic ranging device. Background noise intensity is measured by using a microphone and digital signal processing techniques. The audio output volume is adjusted considering the effect of signal attenuation over user's distance and the effect of background noise. The audio output volume adjustment mechanism can be close-looped, based on the measured signal to noise ratio of acoustic echo of the audio output.