Karaoke - Technology

Technology

A basic karaoke machine consists of a music player, microphone inputs, a means of altering the pitch of the played music, and an audio output. Some low-end machines attempt to provide vocal suppression so that one can feed regular songs into the machine and remove the voice of the original singer; however, this is rarely effective. Most common machines are CD+G, Laser Disc, VCD or DVD players with microphone inputs and an audio mixer built in. CD+G players use a special track called subcode to encode the lyrics and pictures displayed on the screen while other formats natively display both audio and video.

Most karaoke machines have technology that electronically changes the pitch of the music so that amateur singers can choose a key that is appropriate for their vocal range, while maintaining the original tempo of the song. (Old systems which used cassettes changed the pitch by altering playback speed, but none are still on the market, and their commercial use is virtually nonexistent.)

A popular game using karaoke is to type in a random number and call up a song, which participants attempt to sing. In some machines, this game is pre-programmed and may be limited to a genre so that they cannot call up an obscure national anthem that none of the participants can sing. This game has come to be called "Kamikaze Karaoke" or "Karaoke Roulette" in some parts of the United States and Canada.

Many low-end entertainment systems have a karaoke mode that attempts to remove the vocal track from regular audio CDs, using an Out Of Phase Stereo (OOPS) technique. This is done by center channel extraction, which exploits the fact that in most stereo recordings the vocals are in the center. This means that the voice, as part of the music, has equal volume on both stereo channels and no phase difference. To get the quasi-karaoke (mono) track, the left channel of the original audio is subtracted from the right channel. The Sega Saturn also has a "mute vocals" feature that is based on the same principle and is also able to adjust the pitch of the song to match the singer's vocal range.

This crude approach results in the often-poor performance of voice removal. Common effects are hearing the reverberation effects on the voice track (due to stereo reverb on the vocals not being in the center); also, other instruments (snare/bass drum, bass guitar and solo instruments) that happen to be mixed into the center get removed, degrading this approach to hardly more than a gimmick in those devices. Recent years have seen the development of new techniques based on the Fast Fourier Transform. Although still not perfect, the results are usually much better than the old technique, because the stereo left-right comparison can be done on individual frequencies.