An active audio output detection system and method that automatically determines the type of audio output being used on an audio playback device. Embodiments of the system and method include playing back audio content on the audio playback device (such as a mobile phone) and capturing sound during playback using an audio capture device (such as a microphone). The captured sound is analyzed to determine whether the audio playing back on the audio playback device is contained therein. If the captured sound matches the audio playing back on the audio playback device, then the active audio output is a loudspeaker. On the other hand, if the captured sound does not match the audio playing back on the audio playback device, then the active audio output is headphones.