Voice Agent User Identification via Sound Localization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voice-controlled home agent devices struggle to reliably identify users, as speaker recognition alone may not distinguish between users with high probability, especially in situations where multiple users are present or in noisy environments.

Innovation Solution

An apparatus and method that determine a user's location based on sound data and combine it with user identification information from a mobile device, using techniques such as sound localization, beamforming, and machine learning to accurately identify the user, ensuring secure identification even when speaker recognition is imperfect.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If speaker recognition alone is used for user identification, then the system remains simple, but identification reliability deteriorates in noisy environments or when multiple users are present

Engineering Contradiction:
Improveuser identification reliabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines speaker recognition with sound localization technology to create a hybrid identification system. The agent device determines both the speaker's voice characteristics and spatial position, then correlates this with location data from mobile devices to achieve more reliable user identification than speaker recognition alone could provide.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces location information from mobile devices as an intermediary element to enhance user identification. By using the mobile device's GPS or location data as a mediator between the speaker recognition system and the final identification decision, the system achieves higher reliability without directly increasing the complexity of the core recognition algorithm.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If sound localization and mobile device correlation are added to speaker recognition, then user identification accuracy improves, but device complexity increases

Engineering Contradiction:
Improveuser identification accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the user identification process into distinct functional modules: speaker recognition module, sound localization module, and mobile device correlation module. Each module handles a specific aspect of the identification process, allowing the system to achieve high accuracy through coordinated operation of specialized components rather than a single complex system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent makes the agent device multi-functional by integrating voice processing, acoustic spatial analysis, and location data correlation capabilities into a single system. This universal approach allows the device to perform multiple functions (identification, localization, verification) that collectively improve measurement precision while sharing common hardware resources.

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

3Reliability

If multiple identification methods are combined, then reliability in multi-user scenarios improves, but processing time increases

Engineering Contradiction:
Improvemulti-user identification reliabilityVSAvoididentification processing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by continuously tracking and storing the correlation between mobile device locations and registered users before identification is needed. When a user speaks, the system can quickly retrieve pre-computed location-user mappings rather than performing full correlation analysis in real-time, reducing processing time while maintaining high reliability in multi-user scenarios.

Inventive Principle:
Principle #10Preliminary action

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 allows for accurate user identification by correlating sound data with location information from mobile devices, enhancing reliability and security in voice-controlled systems, especially in scenarios where multiple users are present.

Implementation Method 1

determine a location of a user, based on sound data representing speech of the user

Methodology Applied
Scientific EffectSound localization: Acoustics

Data Source

PatentEP3602554B1Apparatus and method
Publication Date: 2023.12.20 SONY GROUP CORP
  • EP3602554B1 patent drawingFigure 1
  • EP3602554B1 patent drawingFigure 2a~3
  • EP3602554B1 patent drawingFigure 4

AI summary

An apparatus has circuitry which determines a location of a user, based on sound data representing speech of the user, and identifies the user based on the determined location of the user and user identification information and location information received from a mobile device.