In-Vehicle Voice User Positioning for Multi-Occupant Interaction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods struggle to accurately locate a specific user in a vehicle when multiple users speak simultaneously, leading to interference and suboptimal interaction experiences.

Innovation Solution

An in-vehicle user positioning method using a microphone array, facial recognition, and mapping relationships between users and locations to determine the user based on voice commands, enabling precise user location even in noisy environments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If voice-based interaction is used to reduce user operations, then ease of operation is improved, but positioning accuracy deteriorates when multiple users speak simultaneously

Engineering Contradiction:
Improveuser operation convenienceVSAvoiduser positioning accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent segments the audio signal into multiple independent sound sources using sound source separation technology. This allows the system to distinguish between multiple users speaking simultaneously and identify which sound source corresponds to the voice command, thereby maintaining positioning accuracy while preserving the ease of voice-based interaction.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary processing layer that includes voice activity detection, sound source separation, and speaker verification modules. This intermediary system processes the raw audio signals before final user identification, enabling accurate positioning even when multiple users are present, thus resolving the contradiction between ease of operation and positioning precision.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Speed

If sound source localization is used to identify user location, then positioning speed is improved, but positioning accuracy deteriorates in noisy in-vehicle environments

Engineering Contradiction:
Improvepositioning speedVSAvoidpositioning accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent merges multiple positioning approaches by combining sound source localization with speaker verification. The system first uses sound source localization to quickly identify potential user locations, then applies speaker verification to confirm the correct user identity. This combination maintains fast positioning speed while improving accuracy in noisy environments by adding a verification layer.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent implements a feedback mechanism where the system continuously monitors audio signals, detects voice commands, verifies speaker identity, and adjusts positioning results accordingly. This feedback loop enables the system to correct positioning errors caused by noise and maintain high accuracy while responding quickly to user commands.

Inventive Principle:
Principle #23Feedback

3Adaptability or versatility

If multiple displays are provided for different users, then adaptability is improved, but user privacy protection deteriorates when the wrong user is located

Engineering Contradiction:
Improvepersonalized service capabilityVSAvoidprivacy protection reliability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent performs preliminary user identification through speaker verification before directing content to specific displays. By verifying the speaker's identity in advance, the system ensures that the correct user profile is activated and the appropriate display is selected, preventing privacy breaches while maintaining the ability to provide personalized services to multiple users.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces mechanical or simple acoustic positioning methods with advanced speaker verification technology that analyzes vocal characteristics. This substitution enables more reliable user identification, ensuring that personalized content is delivered to the correct user and privacy is protected, while still maintaining adaptability to serve multiple users with different preferences.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

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

Accurately identifies the user sending a voice command, allowing for one-to-one interaction and improved user experience by controlling vehicle functions based on the user's location, reducing interference and enhancing personalization.

Implementation Method 1

obtain a sound signal collected by an in-vehicle microphone

Methodology Applied
Scientific EffectSound wave propagation: Sound

Data Source

PatentEP4134949B1In-vehicle user positioning method, in-vehicle interaction method, and vehicle-mounted apparratus
Publication Date: 2025.07.30 YINWANG INTELLIGENT TECHNOLOGIES CO LTD
  • EP4134949B1 patent drawingFigure 1
  • EP4134949B1 patent drawingFigure 2~3
  • EP4134949B1 patent drawingFigure 4

AI summary

This application provides an in-vehicle user positioning method, an in-vehicle interaction method, a vehicle-mounted apparatus, and a vehicle, applied to the field of autonomous driving. The in-vehicle user positioning method includes: obtaining a sound signal collected by an in-vehicle microphone; when a first voice command is recognized from the sound signal, determining a first user who sends the first voice command; and determining an in-vehicle location of the first user based on a mapping relationship between an in-vehicle user and an in-vehicle location. According to the solution of this application, a user who needs to interact can be determined according to a voice command, and then an in-vehicle location of the user who needs to interact can be determined based on the mapping relationship between an in-vehicle user and an in-vehicle location. Even if a plurality of users speak at the same time, the user who needs to interact can be located.