Voice Recognition Controller Using Intermediary Voice Print Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In environments with multiple electronic devices, unintended voice commands can be misrecognized, leading to malfunctions, as voices from one device, such as calling voices or broadcasting sounds, can be mistakenly interpreted by other devices.

Innovation Solution

An electronic apparatus with a voice acquirer, processor, and controller that communicates with other devices to determine whether a received voice is a user's command, preventing operations if it's not intended for it, by comparing voice signals and considering factors like distance and angle of the user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple electronic devices are deployed in the environment to provide comprehensive voice recognition services, then the coverage and accessibility of voice control is improved, but the likelihood of unintended voice command recognition and malfunction increases

Engineering Contradiction:
Improvevoice recognition coverageVSAvoidvoice command accuracy
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces a voice print database as an intermediary that stores and compares voice characteristics. When a voice command is received, the system compares it against stored voice prints to verify authenticity, preventing unintended devices from executing commands. This mediator mechanism resolves the contradiction by enabling reliable identification among multiple devices.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system assigns unique voice print characteristics to each user-device interaction pair, creating localized authentication profiles. Instead of using a single global authentication method, the system tailors voice recognition parameters to specific user-device combinations, improving both coverage and accuracy simultaneously.

Inventive Principle:
Principle #3Local quality

2Speed

If electronic devices continuously monitor and process voice signals to enable responsive voice control, then the responsiveness and user experience is improved, but the risk of misinterpreting non-command voices (such as broadcasting sounds or calling voices) increases

Engineering Contradiction:
Improvevoice control responsivenessVSAvoidvoice command detection accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The system performs preliminary voice print registration and characterization before actual voice commands are given. By pre-storing voice characteristics and establishing baseline patterns, the system can quickly and accurately distinguish between genuine commands and ambient sounds during operation, resolving the speed-precision contradiction.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback mechanisms where voice recognition results are continuously compared against stored voice prints, and the system learns from correct/incorrect recognitions. This feedback loop improves detection accuracy over time while maintaining responsive performance, as the system refines its understanding of legitimate command patterns.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2713351B1Electronic apparatus and control method of the same
Publication Date: 2018.06.20 SAMSUNG ELECTRONICS CO LTD
  • EP2713351B1 patent drawingFigure 1
  • EP2713351B1 patent drawingFigure 2
  • EP2713351B1 patent drawingFigure 3

AI summary

An electronic apparatus includes a voice acquirer which receives a first voice, a voice processor which processes a voice signal, a communication unit which communicates with at least one external electronic apparatus and receives information on at least one second voice, and a controller which determines whether the first voice is a user's command based on the information on at least one second voice transmitted by the communication unit, and if the first voice is not the user's command, does not perform an operation according to the first voice.