A method and apparatus for producing video are provided. The method includes: determining a reference time used as a reference for producing a PIP video; determining a first task time at which to acquire the first image, a second task time at which to acquire the second image, and a third task time at which to acquire the audio; acquiring the first image, the second image, and the audio at the respective task times; and combining the first image, the second image, and the audio according to a result of comparing the reference time and each of the task times, and producing the PIP video. Accordingly, a time and cost for producing the PIP video can be remarkably reduced.