Systems and methods for text tagging and graphical enhancement of subtitles in an audio-visual media display are disclosed. A media asset associated with an audio-visual display that includes one or more speaking characters may be received by a text tagging and graphical enhancement system. A set of sounds from the audio-visual display corresponding to speech by one of the speaking characters is identified. The set of sounds corresponding to the identified speaking character may be analyzed and one or more vocal parameters is identified, each vocal parameter measuring an element of one of the sounds. A display of subtitles synchronized to the speech of the identified speaking character within the audio-visual display may be generated. The appearance of the subtitles may be modified based on the identified vocal parameters for each of the corresponding sounds.