Browse Prior Art Database

Minimizing Power Consumption in Micro-Processor Based Systems which Utilize Speech Recognition Devices

IP.com Disclosure Number: IPCOM000113816D
Original Publication Date: 1994-Oct-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 4 page(s) / 157K

Publishing Venue

IBM

Related People

Cohen, PS: AUTHOR [+2]

Abstract

Described are hardware implementations designed to minimize power consumption in micro-processor based systems which utilize speech recognition devices. The implementation provides multi-state power management and resource allocation techniques for micro-processor based systems, such as portable/laptop type of systems. Five scenarios describe how the technique can be used.

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

Minimizing Power Consumption in Micro-Processor Based Systems which
Utilize Speech Recognition Devices

      Described are hardware implementations designed to minimize
power consumption in micro-processor based systems which utilize
speech recognition devices.  The implementation provides multi-state
power management and resource allocation techniques for
micro-processor based systems, such as portable/laptop type of
systems.  Five scenarios describe how the technique can be used.

      Typically, battery powered micro-processor systems can consume
large amounts of power, particularly in systems which are equipped
with speech recognition capabilities.  This is because power is being
consumed while waiting for the user to say something, as well as
during the implementation of word-spotting functions, where a control
word activates an entire speech-recognition subsystem.  Active
microphones, used in speech recognition systems, not only cause
excessive power consumption, but, in some situations, generate
defective recognition causing the system to perform an incorrect
action.  If a micro-processor based system is unable to sense the
presence of the user, the microphone is being operated needlessly and
inevitably will cause mis-recognition.

      The concept described herein utilizes various
electro-mechanical and optical sensors to activate and de-activate
microphones in speech recognition systems based on the presence and
the appropriate human orientation.  In this way, power consumption
can be reduced.  A multi-state power management and resource
allocation technique is used to distinguish between three states.

The three states are as follows:
  1.  Microphone off.
  2.  Microphone on and waiting for a control word, or control
      phases, as in word-spotting applications.
  3.  A fully active speech-recognition system.
  The following five scenarios illustrate how the power saving
concept
  can be used:
  Scenario #1 - User approaches a kiosk
  o  The microphone is off.  The Disk Access Storage Device (DASD)
     is powered down.  The micro-processor and memory are in their
     lowest power state, capable of monitoring external interrupts.
     The control program, speech-recognition sub-system, and
     initial power state devices of the application are loaded
     and ready to be powered.
  o  The kiosk senses the approach of a user.  This could be by
     way of pressure sensitive pads, infrared detectors, motion
     detectors, push buttons, touch-screen or other electro
     mechanical means of detecting a large object.  The microphone
     is opened, on-screen prompts are displayed, the DASD is spun
     and the entire system is activated.
  Scenario #2 - Word Spotting and Phrase Spotting

      This scenario would be used in most offices, client server, and
home-automation applications and consists of three or more state
solutions where the pres...