Dual Processor Trigger for Low Power Speech Activation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing electronic devices with speech recognition capabilities face high power consumption, particularly in battery-based portable devices, due to continuous power supply to components during standby mode, leading to frequent battery replacement or recharging.

Innovation Solution

An electronic apparatus with a first processor maintaining an active state in low power mode and a second processor in an inactive state, where the first processor sends a trigger signal to activate the second processor upon detecting a sound signal in a specific frequency band, allowing the second processor to perform speech recognition with reduced standby power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If power is continuously supplied to speech recognition components during standby mode, then the speech recognition function can be quickly activated, but power consumption increases

Engineering Contradiction:

Improveactivation speed of speech recognition functionVSAvoidpower consumption during standby

Core Design Contradiction:

SpeedVSUse of energy by moving object

Solution Approach 1:

The system divides the processor into two distinct parts: a first processor that remains active during standby to handle basic operations and detect wake-up triggers, and a second processor that is deactivated during standby but activated upon receiving a trigger signal. This segmentation allows the speech recognition functionality to be quickly activated without continuously powering the entire system, thereby reducing standby power consumption while maintaining fast activation capability.

Inventive Principle:

Principle #1Segmentation

2Speed

If power is continuously supplied to storage components during standby mode, then data can be quickly accessed, but power consumption increases

Engineering Contradiction:

Improvedata access speedVSAvoidpower consumption during standby

Core Design Contradiction:

SpeedVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary actions by having the first processor detect wake-up trigger signals (such as specific frequency band sounds or keywords) during standby mode and prepare trigger signals before the second processor needs to be activated. This preliminary detection and preparation of activation conditions allows the second processor and its associated storage components to be activated only when necessary, reducing unnecessary power consumption while ensuring fast data access when speech recognition is actually needed.

Inventive Principle:

Principle #10Preliminary action

3Use of energy by moving object

If the second processor is completely deactivated during standby, then power consumption is reduced, but activation time increases

Engineering Contradiction:

Improvepower consumption during standbyVSAvoidactivation time of second processor

Core Design Contradiction:

Use of energy by moving objectVSLoss of time

Solution Approach 1:

The first processor serves as an intermediary between the external environment and the second processor. During standby, the first processor remains active and monitors for wake-up triggers. When a trigger is detected, the first processor generates and transmits a trigger signal to activate the second processor. This intermediary role allows the system to maintain low power consumption during standby while ensuring fast activation of the second processor, as the first processor is already running and can immediately initiate the activation sequence upon detecting a trigger.

Inventive Principle:

Principle #24Intermediary (Mediator)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enables quick activation of the speech recognition function while minimizing power consumption, reducing the need for frequent battery replacements or recharging in portable devices.

Implementation Method 1

the first processor, based on sensing a sound signal in a specific frequency band through the microphone in the low power mode

Methodology Applied

Scientific EffectSound detection: Sound

Data Source

PatentEP3677991B1Apparatus and method for transition from low power always listening mode to high power speech recognition mode

Publication Date: 2021.04.21 SAMSUNG ELECTRONICS CO LTD

Figure 1
Figure 2
Figure 3

AI summary

An electronic apparatus is provided. The electronic apparatus includes a microphone, a first processor operatively coupled to the microphone and configured to maintain an active state in a low power mode of the electronic apparatus, and a second processor configured to maintain an inactive state in the low power mode, and the first processor, based on sensing a sound signal in a specific frequency band through the microphone in the low power mode, is configured to control the electronic apparatus to transmit a trigger signal to the second processor to activate the second processor, the second processor including a memory configured to receive power in the low power mode, and configured to be activated based on data stored in the memory based on the trigger signal, and to perform speech recognition regarding the sound signal received from the microphone.