Voice Command Authentication Using Contextual Signals

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Voice-activated devices lack robust security measures to differentiate between authorized and unauthorized users, particularly in scenarios where voice commands can have significant consequences, such as disarming security systems or making large purchases, and are vulnerable to accidental or malicious execution due to shared access.

Innovation Solution

A system that identifies speakers through voice recognition and utilizes contextual signals, such as the presence of a user's mobile device, to authenticate and authorize voice commands, ensuring that only intended actions are executed based on the user's location and device status.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If voice commands are executed without additional authentication, then ease of operation is improved, but security is worsened

Engineering Contradiction:
Improveease of voice command executionVSAvoidsecurity of voice action execution
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system performs preliminary authentication by checking contextual signals (device presence, location, biometric data) before executing voice commands. This preliminary verification ensures that only authorized users can execute sensitive actions, resolving the security concern while maintaining ease of operation for authenticated users

Inventive Principle:
Principle #10Preliminary action

2Reliability

If multiple authentication factors are required, then security is improved, but device complexity is worsened

Engineering Contradiction:
Improvesecurity of voice action executionVSAvoidcomplexity of authentication system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system automatically gathers contextual authentication signals (device presence, location data, biometric information) without requiring explicit user input for each factor. The device self-verified the user's identity by checking multiple contextual signals, providing strong security while maintaining simplicity from the user's perspective

Inventive Principle:
Principle #25Self-service

3Reliability

If contextual signals are verified, then security is improved, but loss of time is worsened

Engineering Contradiction:
Improvesecurity of voice action executionVSAvoidtime for authentication process
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system continuously monitors and caches contextual signals (device presence, location, biometric data) in advance, so that when a voice command is issued, authentication can be performed rapidly by checking pre-collected data rather than gathering all signals from scratch, reducing authentication time while maintaining security

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3671734B1Securely executing voice actions using contextual signals
Publication Date: 2024.05.15 GOOGLE LLC
  • EP3671734B1 patent drawingFigure 1
  • EP3671734B1 patent drawingFigure 2
  • EP3671734B1 patent drawingFigure 3

AI summary

In some implementations, (i) audio data representing a voice command spoken by a speaker and (ii) a speaker identification result indicating that the voice command was spoken by the speaker are obtained. A voice action is selected based at least on a transcription of the audio data. A service provider corresponding to the selected voice action is selected from among a plurality of different service providers. One or more input data types that the selected service provider uses to perform authentication for the selected voice action are identified. A request to perform the selected voice action and (i) one or more values that correspond to the identified one or more input data types are provided to the service provider.