Voice Wakeup Using Multi-Microphone Energy Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current voice wakeup methods for electronic devices require a specific wakeup keyword to be input each time, making the interaction process complex and reducing the human-computer dialogue experience.

Innovation Solution

A voice wakeup method that utilizes at least two microphones distributed in different areas of an electronic device to detect voice energy differences, allowing the device to be woken up without a specific keyword by analyzing the energy of voice signals input into these microphones.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a specific wakeup keyword is required for voice wakeup, then the device can accurately identify when to wake up, but the interaction process becomes complex and user experience deteriorates

Engineering Contradiction:
Improvewakeup identification accuracyVSAvoidinteraction complexity
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent extracts the keyword recognition requirement from the wakeup process, using only voice energy detection instead of requiring specific keywords. This removes the complexity of keyword matching while maintaining reliable wakeup detection through energy threshold comparison.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of detecting presence of specific keywords to trigger wakeup, the patent inverts the approach by detecting absence of keywords combined with sufficient voice energy to trigger wakeup. This reverses the traditional logic from 'keyword present = wakeup' to 'keyword absent but energy sufficient = wakeup'.

Inventive Principle:
Principle #13The other way round (Inversion)

2Reliability

If multiple microphones are used to detect voice energy differences, then wakeup accuracy without keywords is improved, but device complexity increases

Engineering Contradiction:
Improvewakeup detection accuracyVSAvoidmicrophone system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the voice detection function across multiple microphones, with each microphone independently detecting voice energy. The processor then compares energy levels from different microphones to determine wakeup conditions, dividing the complex detection task into simpler individual microphone measurements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The multiple microphones serve dual purposes: they can detect voice energy for wakeup functionality and also provide spatial audio information. This multi-functionality justifies the added hardware complexity by enabling both precise wakeup detection and enhanced audio processing capabilities.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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

Enables a wakeup-keyword-free interaction, enhancing the human-computer dialogue experience by allowing users to wake up devices with any voice input that meets specific energy conditions, improving usability and compliance with natural human interactions.

Implementation Method 1

at least two microphones distributed in different areas of an electronic device to detect voice energy differences

Methodology Applied
Scientific EffectAcoustic to electrical conversion:

Data Source

PatentEP4120266B1Voice wake-up method and device
Publication Date: 2025.06.25 HUAWEI TECH CO LTD
  • EP4120266B1 patent drawingFigure 1
  • EP4120266B1 patent drawingFigure 2
  • EP4120266B1 patent drawingFigure 3~5

AI summary

A speech wakeup method and device, and a readable storage medium are provided. The voice wakeup method includes: detecting voice signals that are input into at least two microphones and that meet a first condition; and determining, based on whether voice energy of the voice signals input into the at least two microphones meets a second condition, whether to wake up the electronic device; and if the second condition is met, waking up the electronic device; or if the second condition is not met, continuing to detect a voice signal input into the microphone. The electronic device can be woken up in a wakeup-keyword-free manner.