Browse Prior Art Database

Method for Appropriately Interfacing to User Characteristics in a Voice Interface System

IP.com Disclosure Number: IPCOM000111656D
Original Publication Date: 1994-Mar-01
Included in the Prior Art Database: 2005-Mar-26
Document File: 2 page(s) / 37K

Publishing Venue

IBM

Related People

Haynes, TR: AUTHOR

Abstract

It is desirable to interface to a user so that the user enjoys using a computer. Proliferation of voice control systems sometimes use and assume a single voice in order to interface with the system. A particular voice is used to prompt and interface with the user. A method is desirable for using a voice which is most compatible with the individual nuances inherent in the user's own voice.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 87% of the total text.

Method for Appropriately Interfacing to User Characteristics in a
Voice Interface System

      It is desirable to interface to a user so that the user enjoys
using a computer.  Proliferation of voice control systems sometimes
use and assume a single voice in order to interface with the system.
A particular voice is used to prompt and interface with the user.  A
method is desirable for using a voice which is most compatible with
the individual nuances inherent in the user's own voice.

      This idea automatically determines the best voice to use on a
system which talks to a user for interfacing with the user.  Provided
is a voice recognition algorithm which determines the rate and volume
of the user's speech pattern.  The system then adjusts its
computer-generated voice to match.  This capability can be very
useful for certain users.  For instance, a user with hearing
difficulties often compensates by speaking louder than normal.  It
would benefit such a person if the system would speak louder than
normal, without forcing the user to explicitly adjust the speech
output volume themselves.  Also, if a user speaks more slowly than
normal, it might be helpful if the system voice adjusted its rate
down as well.  Of course, both of these voice characteristics should
be under user control for manual override.

The essential Elements are:

o   Voice recognition criteria mapped to synthesized voice.

o   Selection of the best matching criteria to the user's voice...