A recording medium and method and apparatus for reproducing and recording text subtitle streams are disclosed. Each text subtitle stream includes a dialog style segment defining a set of region styles and at least one dialog presentation segment, each of which contains one or more regions of dialog text. Each region of dialog text is linked to one of the set of region styles defined in the dialog style segment, and includes at least one pair of an inline style and a text string. The inline style is configured to change one of region presentation properties specified by the linked region style for the text string.