First ratio acquisition means (102) of a display timing determination device (10, 20) acquires, for each voice, a first ratio of an interval between output timings of voices that are different in output order by a first order difference to an interval between output timings of voices that are different in output order by a second order difference. Second ratio acquisition means (104) acquires, for each piece of character information, a second ratio of an interval between provisional display timings of pieces of character information that are different in display order by the first order difference or a third order difference corresponding to the first order difference to an interval between provisional display timings of pieces of character information that are different in display order by the second order difference or a fourth order difference corresponding to the second order difference. Identification means (105) identifies, based on the first ratio of each voice and the second ratio of each piece of character information, an association relationship between the voice and the piece of character information. Display timing determination means (106) determines a definitive display timing of each piece of character information based on the association relationship.