Another technology format for presenting captioning along with music is Karaoke. Karaoke is basically a collection of multi-media devices intended to support sing-along performances. This adapted use of captioning is quite popular in nightclubs in Japan, where singing into a microphone has become extremely fashionable.

            Karaoke uses prerecorded, professional musical accompaniments and video appropriate to the song while sing-along words are scrolled at the bottom of the picture. The text changes color with the rhythm and speed of the melody, allowing even a novice reader to track text-sound and phoneme relationship (Kothari & Takeda, 2000). There is a previously recorded vocal, which is kept on a separate audio track from the accompaniment, so it can either be sang along with or faded out. Karaoke videos are visually appealing, and the songs are usually contemporary popular vocals. One of the motivating factors of this device is that students are often surprised by how good their singing voices sound when backed up by electronic accompaniment. They are often amazed at how good they sound with backup.

      M.J. Wagner and J.S. Brick (1993) demonstrated that karaoke can be a powerful reinforcer to what is basically a repetitive reading activity. In that study, students also indicated that it was easier to learn the words to the songs via the karaoke because the songs were visually related to the video. Further, the students felt that the use of a microphone dramatically increased their focus and engagement. Wagner and Brick felt that the absence of sheet music enabled the students to focus all of their attention on the screen and the lyrics.

      Karaoke does have some problems when used as a reading activity. With Karaoke, most available lyrics are suitable for instruction at the below the fourth grade level, and it may be difficult to find music to more challenging material. Second, commercially- available-Karaoke typically uses an audio source that is less dynamic and less clear than original sources; the vocal model is often faded to the background and the tempo and melody are often altered or reduced.