Browse Prior Art Database

Postal Scale with Speech Recognition Feature

IP.com Disclosure Number: IPCOM000114016D
Original Publication Date: 1994-Oct-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 2 page(s) / 86K

Publishing Venue

IBM

Related People

Cohen, PS: AUTHOR [+3]

Abstract

Disclosed is a postal scale having a speech recognition feature, operating with a speaker-independent, continuous-speech recognition system, which can be used immediately after the user walks up to the device, such as the IBM ICSS (IBM Continuous Speech Series), to control basic scale functions. The scale may be applied in the context of a postal kiosk, including means to dispense stamps or provide imprints from a postage meter. Additionally, speech recognition may be incorporated into other types of scales, such as bathroom scales, meat-counter scales, and truck-stop scales.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Postal Scale with Speech Recognition Feature

      Disclosed is a postal scale having a speech recognition
feature, operating with a speaker-independent, continuous-speech
recognition system, which can be used immediately after the user
walks up to the device, such as the IBM ICSS (IBM Continuous Speech
Series), to control basic scale functions.  The scale may be applied
in the context of a postal kiosk, including means to dispense stamps
or provide imprints from a postage meter.  Additionally, speech
recognition may be incorporated into other types of scales, such as
bathroom scales, meat-counter scales, and truck-stop scales.

      The maximum branching factor is constrained by limiting the
active vocabulary and active phrases, increasing the accuracy of
speech recognition while reducing the need for computation.  This may
be achieved, for example, by providing the output from the scale as
input to the speech-recognition system.  limiting the words, phrases,
grammars, etc., active in the speech-recognition system.  The active
questions, grammars, or contexts, including words, sentences, and
phrases grouped into context, which may be used to formulate
acceptable questions or commands, may then be displayed on a screen.

      In general, a measuring device, such as a scale or other
pressure-sensing instrument, an optical sensor, or a motion detector,
provides input to the speech-recognition system, causing it to turn
on and off, and causing the active vocabularies, grammars, contexts,
n-grams or language models to be changed or constrained.

      In the specific example of a postal kiosk, the user walks up to
the kiosk, which is activated by touching a pressure-sensitive pad,
by a telephone handset being picked up, or by a package or letter
being placed on the scale.  Once the kiosk is activated, the
speech-recognition system selects a context, such as a set of words,
phrases, commands, etc., which it is prepared to hear.  Preferably,
the kiosk displays prompts on the screen such as, "Say one of the
following--- find a ZIP code, find overseas packages and letters, or
overnight delivery."

      The user might then place a package on the scale, causing the
display of a new message...