Contents

Karaoke lets people sing along to recorded accompaniments. Many karaoke systems provide the accompaniment, on-screen vocal texts for the singer to follow, and a video that is appropriate to the selected song. There are currently three types of karaoke systems: those for karaoke houses, cable TV companies, and home use. These are all standalone systems that require considerable human interaction. None of them can provide Karaoke on demand. To provide KOD, Y.-J. Lee, D.H.C. Du, and W.-H. Ma (1996) have proposed Sesame (a scalable and extensible architecture for multimedia entertainment). They claim that Sesame is modular (system elements can be easily replaced), scalable, extensible, and cost effective. Although other researchers have devised KOD schemes, these schemes are also standalone systems, are limited by throughput, or only provide audio. Sesame system would have three basic components: the server system, the network, and the user workstations. The server system uses off the shelf products such as a Fast-Wide SCSI II channel to connect storage devices to servers. Servers, also using commodity products such as Intel Pentium or Motorola PowerPC chips, administer the data. A switch based network connects the server to the workstations. A switch based network can provide each user with a dedicated link and bandwidth at low latency, unlike a shared medium network such as a LAN. The workstation forwards user requests, decompresses the data, mixes the user´s voice with the soundtrack, and controls the audio-video display. The system uses MPEG compression.