Browse Prior Art Database

Method and System For Automatic Karaoke Video Reproduction

IP.com Disclosure Number: IPCOM000125668D
Original Publication Date: 2005-Jun-10
Included in the Prior Art Database: 2005-Jun-10
Document File: 5 page(s) / 194K

Publishing Venue

IBM

Abstract

Disclosed is a method and system that can automatically produce a new Karaoke video from an existing one. Specifically, the new video contains the original lyrics and soundtrack extracted from the existing Karaoke video which are critical elements to a Karaoke video. It, however, has a different background video scene which can be either a home video, a movie clip or any video that contains interesting content and are preferred by the customer or Karaoke production company.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 49% of the total text.

Page 1 of 5

Method and System For Automatic Karaoke Video Reproduction

KTV, which stands for Karaoke TV, has become increasingly popular in Asian countries in recent years. Unlike MTV (Music TV) where the artist performs the singing and audience passively watch the video, KTV allows audience to sing along with the music. Moreover, considering that the amateur singer may not be able to remember all song words , almost all KTV provides lyrics which are superimposed onto video frames in form of closed captions and displayed at the exact moment when they are sung. In this context, different colors are used to display different texts in order to remind audience of their "status". A text character is in an active status if it is being sung; otherwise, it is inactive. Figure 1 shows one example where there are two text lines in the image taken from a Chinese KTV video. Specifically, notice that texts in the second line are all white, while the leading texts in the first line are blue. This implies that the first line of texts are the song words just or being sung, while the second line is still waiting to be activated. Moreover, as the audience sing along with the music, the inactive words become active and their color will turn to blue in a fairly synchronized way. Obviously, this change-of-text-color mechanism can natually and effortlessly guide audience to complete a song which they may not be familiar with.

Figure 1: A frame taken from a Chinese KTV video

It is unsurprising that KTV has gained (and is continuously gaining) a lot of popularity since it provides a wonderful way for people to learn a new song, to entertain themselves, or to share a happy time with family and friends. Moreover, all of these could be enjoyed at a rather low cost. All needed is a microphone, a TV, and a Karaoke CD that contains the desired song.

However, the production of a Karaoke video is not cheap. In fact, it is rather expensive. To make its content attractive, a special production team is needed which selects cast, shoots video, and mixes them with the song to best interpret the music theme. Nevertheless, different people may have different tastes and preferences. Thus very often, for every single song, one can easily find multiple Karaoke videos on the market. This obviously boost the production expenses. In this context, a better approach is needed which can not only make Karaoke video customized to people's need, but also

1

[This page contains 8 pictures or other non-text objects]

Page 2 of 5

keep the production cost low at the same time.

This invention proposes a system and a method that automatically produce a new Karaoke video by mixing the lyrics and soundtrack extracted from an existing Karaoke video with a new background video. Specifically, assume that video V contains the desired music- and time- synchronized lyrics L, while video C contains the desired new background scene, which could be either a home video, a movie clip, or any video that contains i...