Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Computer-Aided Dialog Synchronization and Subtitle Generation for Video Systems

IP.com Disclosure Number: IPCOM000105473D
Original Publication Date: 1993-Aug-01
Included in the Prior Art Database: 2005-Mar-19
Document File: 2 page(s) / 57K

Publishing Venue

IBM

Related People

Bandara, U: AUTHOR [+3]

Abstract

Described is a system in which the speech signal track in a video system is subjected to speech recognition and then to machine trans lation. After the operating delay has been compensated for, the image, the speech and the underlying text are synchronized.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 68% of the total text.

Computer-Aided Dialog Synchronization and Subtitle Generation for Video Systems

      Described is a system in which the speech signal track in a
video system is subjected to speech recognition and then to machine
trans lation.  After the operating delay has been compensated for,
the image, the speech and the underlying text are synchronized.

      An internal automatic speech recognizer is implemented in a
video system such that the speech signal in the audio channel can be
processed by means of the recognizer to produce the underlying
orthographic text.

      The speech signal is buffered in advance so that the operating
delay of the recognizer is compensated for to synchronize the three
channels (the graphic (image), the subtitle text, and the audible
speech signal (Fig. 1)).  For this purpose, the sound track
transducer precedes the graphic (image) track with respect to the
movement of storage media (magnetic tape) with a constant time t3.
Then, the speech signal is divided into two parts, one part being
sent to a signal buffer and the other part being sent through the
speech recognizer and, optionally and subsequently, through the
machine translator.  A buffering device is provided both for the
output text of the recognizer or translator and the direct speech
signal such that the text and the associated speech signal are
simultaneously released from the buffer after a time t1 measured from
the moment the signal is recognized and/or translated.  The
ap...