In-Vehicle Voice Separation for Single-Receiver Cabin Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The high cost of configuring multiple voice receivers in vehicles for voice recognition capabilities, which increases the overall cost and data processing complexity.

Innovation Solution

A method that separates initial voice data received from multiple regions within a vehicle into sub-data and corresponding description information, allowing a single voice receiver to determine the vehicle's working mode based on the sub-data, thereby reducing the number of required voice receivers and improving processing performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple voice receivers are configured in different regions of the vehicle, then voice recognition coverage is improved, but device cost and data processing complexity increase

Engineering Contradiction:
Improvevoice recognition coverageVSAvoiddata processing complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

A single voice receiver is designed to perform multiple functions by receiving and processing voice data from multiple vehicle regions. The receiver uses beamforming technology to identify voice sources from different directions (driver, front passenger, rear passengers) and determines working modes based on the number and location of voice sources, eliminating the need for separate receivers in each region while maintaining comprehensive voice recognition coverage.

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

2Adaptability or versatility

If multiple voice receivers are configured in different regions of the vehicle, then voice recognition coverage is improved, but device cost increases

Engineering Contradiction:
Improvevoice recognition coverageVSAvoidnumber of voice receivers
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

A single voice receiver is designed to perform multiple functions by receiving and processing voice data from multiple vehicle regions. The receiver uses beamforming technology to identify voice sources from different directions (driver, front passenger, rear passengers) and determines working modes based on the number and location of voice sources, eliminating the need for separate receivers in each region while maintaining comprehensive voice recognition coverage.

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

Solution Approach 2:

Multiple voice reception functions that would traditionally require separate receivers are merged into a single receiver. The receiver combines signal processing capabilities to handle voices from multiple regions simultaneously, using beamforming to spatially separate and identify different voice sources, thereby reducing the total number of receivers from multiple to one while maintaining full coverage.

Inventive Principle:
Principle #5Merging (Combining)

3Device complexity

If a single voice receiver is used to process voice data from multiple regions, then cost and processing load are reduced, but voice recognition accuracy may deteriorate

Engineering Contradiction:
Improvedata processing complexityVSAvoidvoice recognition accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces traditional mechanical signal processing methods with beamforming technology, which uses signal processing algorithms to spatially filter and identify voice sources. The beamforming process calculates direction of arrival (DOA) of voice signals and determines the number of voice sources through spectral analysis, enabling accurate identification of voice regions without requiring multiple physical receivers. This substitution maintains high recognition accuracy while reducing system complexity.

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

Data Source

PatentEP4030424B1Method and apparatus of processing voice for vehicle, electronic device and medium
Publication Date: 2024.02.07 APOLLO INTELLIGENT CONNECTIVITY (BEIJING) TECH CO LTD
  • EP4030424B1 patent drawingFigure 1
  • EP4030424B1 patent drawingFigure 2
  • EP4030424B1 patent drawingFigure 3

AI summary

A method and an apparatus of processing a voice for a vehicle, a device, a medium and a product are provided, which relate to a field of voice recognition technology. The method of processing a voice for a vehicle includes: separating an initial voice data in response to receiving the initial voice data from a plurality of regions inside the vehicle, so as to obtain a plurality of voice sub-data and a description information for each voice sub-data of the plurality of voice sub-data, the plurality of voice sub-data correspond to the plurality of regions respectively, and the description information for each voice sub-data indicates the region corresponding to the each voice sub-data in the plurality of regions; and determining a voice working mode of the vehicle based on the plurality of voice sub-data.