Systems, devices, and methods for voice over music recording are provided. A person may sing along with music being output into the physical environment of the person by a loudspeaker. One or more microphones may be used to record both the voice of the person singing, and the music output by the speaker. An electronic device may cancel the portion of the microphone signals containing the music output by the speaker, and remix the voice of the person with digital audio content corresponding to the music.