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

Speech Recognition Methods for Controlling Cable Television

IP.com Disclosure Number: IPCOM000116184D
Original Publication Date: 1995-Aug-01
Included in the Prior Art Database: 2005-Mar-30
Document File: 4 page(s) / 157K

Publishing Venue

IBM

Related People

Cohen, PS: AUTHOR [+2]

Abstract

Disclosed are methods for using speech recognition to select or modify images, sound, and data transmitted on a cable television system. A first method connects the public telephone network with a speech recognition system and with the cable television network. A second method uses the cable television network to transmit speech commands to a speech recognition system.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 42% of the total text.

Speech Recognition Methods for Controlling Cable Television

      Disclosed are methods for using speech recognition to select or
modify images, sound, and data transmitted on a cable television
system.  A first method connects the public telephone network with a
speech recognition system and with the cable television network.  A
second method uses the cable television network to transmit speech
commands to a speech recognition system.

      Fig. 1 is a diagram of a system in which the first method is
applied.  A user's house 1 includes a conventional television set 2
connected to a cable television line 3.  An optional cable box 4 may
be used for this connection.  A conventional telephone 5 is connected
to a conventional telephone network 6.  At a remote location the
switching system 7 of the telephone company is connected to a speech
recognition system 8, which is in turn connected to a cable
television signal generator 9.  A standard cable television signal,
brought into the signal generator  on cable 10, is modified within
the signal generator, according to instructions from speech
recognition system 8, to form the modified signal transmitted to the
user on line 3.

      The user begins the process of using the system of Fig. 1 by
placing a call to the cable television system with telephone 5.  The
cable system identifies the user with, for example, the Automatic
Number Identification (ANI) protocol, or asks the user to provide his
phone number or password by speech or by depressing a sequence of
telephone buttons.  The cable system them displays a series of
prompts on the screen of television set 2, on a specific channel or
overlaid on all channels.  This process may cause the existing
channel window to shrink, providing a split-screen mode with the
prompts displayed alongside the normal picture received on the
channel.  The cable system can additionally transmit audio messages
to the user through the telephone network.  This capability is used,
for example, to allow the user to listen to an alternative audio
broadcast corresponding to the channel being viewed.  Voice commands,
interpreted through speech recognition system 8, provide for various
functions, such as changing channels, identifying a subset of active
channels through which the user may move, viewing multiple channels
which are windowed or tiled, displaying text on the screen, selecting
answers to multiple-choice questions, seeing scores or personal game
totals displayed on the screen, or selecting movies to review.

      Within speech recognition system 8, the active words, phrases,
n- grams, and grammars are constrained to a minimum number sufficient
to support a given screen, panel, or set of prompts.  The active
context is dynamically changed by user and program input.  The
resulting images, sounds, and prompts are displayed on the screen
means of techniques such as compression, individually addressed
television sets, and signaling the...