Voice Wakeup Co-processor for Low Power Consumption

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voice control systems in electrical appliances face high power consumption due to continuous operation of voice recognition processors, as they are activated regardless of the presence of human voice, leading to inefficient energy use.

Innovation Solution

A wakeup method and apparatus that collect and process voice information to recognize specific wakeup words, activating the voice recognition processor only when a human voice is detected, using a co-processor for initial voice processing and blind-source separation to reduce unnecessary operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If the voice recognition processor is activated for real-time monitoring of external sounds, then the voice control function is available, but the power consumption increases

Engineering Contradiction:
Improvevoice control availabilityVSAvoidpower consumption
Core Design Contradiction:
Ease of operationVSUse of energy by moving object

Solution Approach 1:

The voice processing system is divided into two segments: a low-power co-processor that performs initial voice activity detection and basic processing, and a full-featured voice recognition processor that only activates when necessary. This segmentation allows the system to maintain voice control availability while minimizing power consumption during idle periods.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The co-processor performs preliminary voice processing actions before the main voice recognition processor needs to activate. By detecting voice activity and pre-processing signals in advance, the system reduces the frequency and duration of full processor activation, thereby reducing overall power consumption while maintaining responsiveness.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If the voice recognition processor keeps working for real-time monitoring, then voice recognition accuracy is maintained, but energy consumption increases

Engineering Contradiction:
Improvevoice recognition accuracyVSAvoidenergy consumption
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

Instead of continuous operation, the system uses periodic action by having the co-processor monitor voice activity at intervals and only activating the full voice recognition processor when voice activity is detected. This periodic monitoring approach maintains recognition accuracy while dramatically reducing energy consumption compared to continuous operation.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The co-processor serves as an intermediary between the ambient environment and the main voice recognition processor. It filters and pre-processes audio signals, acting as a gatekeeper that determines when the energy-intensive main processor should activate, thus maintaining accuracy while minimizing energy loss.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Speed

If the voice recognition processor is always active, then response time is reduced, but power consumption increases

Engineering Contradiction:
Improveresponse timeVSAvoidpower consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The co-processor performs preliminary voice activity detection continuously at low power, so when actual voice input occurs, the main processor can activate immediately without detection delay. This preliminary monitoring maintains fast response time while avoiding the energy cost of keeping the full processor continuously active.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The co-processor acts as an intermediary that maintains system responsiveness without requiring the main processor to be always active. It quickly detects voice events and triggers the main processor only when needed, achieving fast response times with minimal energy consumption.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3522153B1Voice control system, wakeup method and wakeup apparatus therefor, electrical appliance and co-processor
Publication Date: 2023.12.27 HEFEI HUALING CO LTD
  • EP3522153B1 patent drawingFigure 1~2
  • EP3522153B1 patent drawingFigure 3~4
  • EP3522153B1 patent drawingFigure 5

AI summary

A voice control system, a wakeup method and wakeup apparatus therefor, an electrical appliance and a co-processor. The wakeup method comprises: a collection step for collecting voice information; a processing step: processing the voice information to determine whether the voice information contains a human voice; if so, separating a voice information segment containing the human voice, and entering a recognition step; the recognition step: performing wakeup word recognition on the voice information segment containing the human voice; if a wakeup word is recognised, then entering a wakeup step; and if no wakeup word is recognised, then returning to the collection step; and the wakeup step: waking up a voice recognition processor. Each part is designed in a modularized manner according to the method. The voice recognition processor only operates when voice recognition is required, avoiding ceaseless all-weather operation, and having reduced energy consumption. The voice wakeup apparatus only recognises a wakeup word, has low power consumption, and consumes very little energy even in all-weather operation, solving the problem of high power consumption in existing voice recognition.