Multi-Device Voice Response Selection via Direct Sound Energy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In a multi-device voice interaction system, multiple electronic devices with sound pickup functions can be inadvertently activated by a single wake-up word, leading to confusing results and reduced user experience due to simultaneous responses.

Innovation Solution

A method that determines a target sound source point and direct sound energy for each electronic device, allowing only devices with sufficient direct sound energy to respond, thereby selectively activating the appropriate devices based on the user's voice command.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If multiple electronic devices use the same wake-up word for voice interaction, then the system becomes more accessible and easy to use, but multiple devices may be inadvertently activated simultaneously causing confusing results

Engineering Contradiction:
Improvevoice interaction accessibilityVSAvoidresponse accuracy
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent applies local quality by making each electronic device have distinct acoustic characteristics through unique device identifiers embedded in their audio responses. When a wake-up word is detected, each device calculates its own direct sound energy and compares it with others, allowing the system to locally determine which device should respond based on acoustic proximity rather than universal activation. This resolves the contradiction by maintaining ease of operation through wake-up word recognition while improving reliability through device-specific response selection.

Inventive Principle:
Principle #3Local quality

2Adaptability or versatility

If all electronic devices with sound pickup functions respond to the same wake-up word, then the system provides comprehensive voice interaction coverage, but the sound pickup results become confusing and user experience deteriorates

Engineering Contradiction:
Improvevoice interaction coverageVSAvoidsound pickup accuracy
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies inversion by reversing the traditional approach where all devices respond to a wake-up word. Instead, when a wake-up word is detected by multiple devices, the system inverts the selection logic by having each device calculate its direct sound energy and select only the device with the highest value to respond. This maintains comprehensive voice interaction coverage through wake-up word detection while improving sound pickup accuracy by selecting a single appropriate device based on inverted selection criteria.

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

3Measurement precision

If electronic devices calculate direct sound energy to determine which device should respond, then the response accuracy improves, but the computational complexity and processing time increase

Engineering Contradiction:
Improvesound source localization accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies partial action by having devices perform direct sound energy calculation only when necessary - specifically when multiple devices detect the same wake-up word. In normal operation, devices use simple wake-up word detection. The complex direct sound energy calculation is performed partially only for devices that need to determine response priority, reducing overall processing complexity while maintaining high measurement precision for sound source localization when needed.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3783604B1Method for responding to voice signal, electronic device, medium and system
Publication Date: 2024.03.27 BEIJING XIAOMI INTELLIGENT TECH CO LTD
  • EP3783604B1 patent drawingFigure 1~2
  • EP3783604B1 patent drawingFigure 3~4
  • EP3783604B1 patent drawingFigure 5

AI summary

The present disclosure relates to a method, an electronic device, a medium and a system for responding to a voice signal, the method including: receiving (S11) a voice signal by the plurality of electronic devices; determining (S12), for each of the electronic devices, a target sound source point associated with the voice signal; determining (S13), for each of the electronic devices, a direct sound energy associated with the voice signal according to the target sound source point; and selecting (S14) from the plurality of electronic devices at least one electronic device with a direct sound energy satisfying a predetermined condition, and responding to the voice signal by the at least one electronic device. It realizes a sound pickup decision system based on a plurality of electronic devices, which can make more accurate, more reasonable, and more targeted response to a user's voice, improving voice interaction experience.