An after recording method comprising a first step of reading the original data including at least a video bit stream from a digital recording medium (1), a second step of decoding the original data, a third step of encoding audio data in synchronism with the decoding of the original data thereby to form a new audio bit stream, and a fourth step of writing the new audio bit stream in an area of the digital recording medium 1 corresponding to a time point substantially equal to the recording area of the original data. The sound after-recorded can be easily reproduced in accurate temporal agreement with the original image.