A method for setting a repeated reproduction section and for controlling its execution is disclosed, in which when learning a language or a music, the user of a digital audio player can accurately and efficiently set a starting point and an ending point of the repeated reproduction section, so that any continuous loss of words or clauses can be essentially prevented. The method for setting a repeated reproduction section and for controlling its execution includes the following steps. A language or song learning digital audio data file is distinguished and mapped into section information units (first step). A starting point of a currently reproduced section information unit is set as the starting point for the repeated reproduction if there is an input of a starting point setting key for the repeated reproduction during a reproduction of the digital audio data of the first step (second step). An ending point for the currently reproduced section information unit is set as an ending point for the repeated reproduction if there is an input of an ending point setting and executing key for the repeated reproduction, so as to decide the repeated reproduction section and so as to directly execute the repeated reproduction (third step).