Browse Prior Art Database

Automation of captioning during live performance Disclosure Number: IPCOM000124863D
Original Publication Date: 2005-May-10
Included in the Prior Art Database: 2005-May-10
Document File: 3 page(s) / 80K

Publishing Venue



A methodology for automation of captioning during a live performance utilizing voice recognition and computer audio processing.

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 52% of the total text.

Page 1 of 3

Automation of captioning during live performance

Disclosed is a methodology for the automation of captioning during a live performance. Live performances today often have the dialog or lyrics displayed above the stage or transmitted to a device such as a PDA (Personal Digital Assistant) to permit those with difficulty in hearing to follow the dialogue. Current solutions are typically a manual process, involving a person running a projector, changing the projected dialog at the proper time to coincide with the stage action. The problem with current solutions is that by being manual they are rife for error, given the duration and intensity of the performance. Current solutions also tend to be difficult for dialogue that does not have a specific tempo, such as live plays (Shakespeare, for example). The methodology disclosed eliminates human control and replaces it through the use of computer software, ensuring greater accuracy and ability to expand from current usage potentially into new areas of performance.

The process in this methodology is to automate the display of the dialog or action by capturing audio from the performance space through the use of microphones located either in the space or on the performers. Once the audio has been received by the computer software, pattern recognition and pseudo-AI algorithms are used to match the progression of the performance through context sensitivity and word recognition and display corresponding text to the appropriate medium. This eliminates the requirement for a person to manually perform these actions, as well as improving overall accuracy in aligning the projection with the performance. The projection could be based upon a scripted version of the material, or on transliteration of the actual dialog which, presuming very accurate pattern recognition, would take into account variations in each performance.

A sample implementation of this methodology can be broken into two sections: Pre-performance and Performance. It is in the combination of the two sections that generates the differentiation of this methodology.


1. The anticipated dialogue that is to be input into the computer software.

     The first step of the methodology is to input the dialogue that is to anticipated to be orated during the production into the computer software...