Bone Conduction Voice Registration for Real-Time Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing voice extraction techniques face challenges in achieving real-time voice extraction for any target speaker, particularly due to the need for pre-registration in quiet environments and the difficulty in handling changes in the user's voice feature.

Innovation Solution

The method involves determining registered audio information associated with a user based on a bone conduction (BC) signal, extracting a second audio signal corresponding to the user from a first audio signal, and processing this information to improve voice extraction quality without the need for advance registration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If pre-registration in quiet environments is used, then voice extraction quality is improved, but device complexity and ease of operation deteriorate

Engineering Contradiction:

Improvevoice extraction qualityVSAvoidoperation convenience

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system performs preliminary voice feature registration by capturing bone conduction signals during initial use, storing these features for later real-time extraction. This preliminary action eliminates the need for separate quiet environment registration sessions, improving ease of operation while maintaining extraction quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system automatically captures and registers voice features during normal usage through bone conduction signals, without requiring user intervention to find quiet environments or perform separate registration steps. The device serves itself by continuously learning and updating voice features in real-world conditions.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If pre-registration in quiet environments is used, then voice extraction quality is improved, but loss of time increases

Engineering Contradiction:

Improvevoice extraction qualityVSAvoidregistration time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system continuously captures bone conduction signals and updates voice features during normal usage rather than requiring a separate registration phase. This continuous learning process eliminates time loss by integrating feature capture into the actual usage workflow, maintaining extraction quality without dedicated registration time.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

Voice features are registered preliminarily during the first few uses through bone conduction signal capture, enabling immediate real-time extraction without requiring extended quiet environment sessions. This preliminary registration during natural usage significantly reduces the time investment required.

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If traditional voice extraction methods are used, then adaptability to voice feature changes deteriorates, but device complexity is reduced

Engineering Contradiction:

Improveadaptability to voice feature changesVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system continuously monitors bone conduction signals and compares them against registered voice features, detecting changes in real-time. When voice feature changes are detected, the system automatically updates the registered features, creating a feedback loop that adapts to user changes without requiring manual re-registration or complex manual adjustment mechanisms.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The voice feature registration is made dynamic rather than static - the system continuously updates registered features based on current bone conduction signals. This dynamic adaptation allows the system to handle voice feature changes (such as illness, aging, or intentional modification) while maintaining manageable complexity through automated processes.

Inventive Principle:

Principle #15Dynamics

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 enables adaptive real-time registration and updating of the user's voice, improving the quality of voice extraction and naturalness of the voice, even when the user's voice feature changes.

Implementation Method 1

determining registered audio information associated with a user based on a bone conduction (BC) signal

Methodology Applied

Scientific EffectBone conduction:

Data Source

PatentUS20250029616A1Method and apparatus for registering and updating audio information associated with a user

Publication Date: 2025.01.23 SAMSUNG ELECTRONICS CO LTD

AI summary

According to an embodiment of the disclosure, a method may include determining registered audio information associated with a user based on a bone conduction (BC) signal. According to the embodiment of the disclosure, the method may include extracting a second audio signal corresponding to the user from a first audio signal based on the registered audio information associated with the user. According to the embodiment of the disclosure, the method may include processing the at least one from among the extracted second audio signal and a portion of the first audio signal which does not contain the second audio signal.