Voiceprint Recognition Model Updating for Wake-Free User Activation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voiceprint recognition systems face challenges in accurately recognizing user voices due to variations caused by environmental, psychological, pathological, and age-related factors, leading to inconsistent recognition performance.

Innovation Solution

The system trains a voiceprint model on a user's voice to create an exclusive model, allowing for accurate recognition by updating the model based on the user's voice features in different states, including emotional and speech speed variations, and uses synthesized voices for additional training data to enhance model accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If voiceprint recognition is performed using traditional wake word activation, then the device remains in a low-power state most of the time, but the user must remember and speak a specific wake word which adds operational complexity

Engineering Contradiction:
Improveease of activationVSAvoidrecognition process complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent extracts and removes the wake word requirement from the voice assistant activation process. The system continuously monitors voice inputs in the background without requiring any specific trigger phrase, allowing users to simply speak naturally to activate functionality.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The voiceprint recognition system performs self-identification and activation without external triggers. The processor automatically detects and processes voice inputs, comparing them against stored voiceprint templates to determine when activation is needed, eliminating the need for user-initiated wake words.

Inventive Principle:
Principle #25Self-service

2Speed

If the processor continuously monitors voice inputs for wake words, then activation becomes more responsive, but power consumption increases significantly

Engineering Contradiction:
Improveactivation response speedVSAvoidprocessor power consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The voice processing system is segmented into multiple stages: a low-power initial filtering stage that detects potential voice inputs, followed by a higher-power voiceprint comparison stage only when activation is detected. This hierarchical approach maintains responsiveness while minimizing overall power consumption.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system employs periodic voice sampling at low power intervals, with full voiceprint analysis performed only periodically when activation conditions are met. This allows the system to maintain awareness of the acoustic environment without continuously consuming high power.

Inventive Principle:
Principle #19Periodic action

3Reliability

If voiceprint templates are stored securely in encrypted form, then user privacy and security are protected, but the verification process becomes more complex and time-consuming

Engineering Contradiction:
Improvevoice data securityVSAvoidverification time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Voiceprint templates are pre-processed and stored in an optimized encrypted format during device setup. The encryption scheme is designed in advance to allow for efficient comparison operations, so that during actual verification, the system can quickly perform matching without extensive real-time computation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system transforms the voiceprint data into a specific parameter representation that balances security and speed. By changing the mathematical parameters of how voiceprints are stored and compared (using efficient cryptographic hashing and feature extraction), the system achieves both strong security and fast verification.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4293664B1Voiceprint recognition method, graphical interface, and electronic device
Publication Date: 2026.05.20 HONOR DEVICE CO LTD
  • EP4293664B1 patent drawingFigure 1
  • EP4293664B1 patent drawingFigure 2
  • EP4293664B1 patent drawingFigure 3A

AI summary

This application discloses a voiceprint recognition method, a graphical interface, and an electronic device. In the voiceprint recognition method, a voiceprint model is preset in the electronic device, and then the electronic device trains and updates the preset voiceprint model based on a voiceprint feature extracted from a voice of a registered user to obtain an exclusive voiceprint model belonging to the registered user. Finally, the electronic device uses the exclusive voiceprint model to generate a registered user representation based on the voiceprint feature of the voice of the registered user, and uses the registered user representation as a reference standard to realize voiceprint recognition on a voice of a speaker. Since the exclusive voiceprint model is trained based on voiceprint features of personal voices of the registered user, the registered user representation generated can accurately express voiceprint features of the user, thereby improving accuracy of voiceprint recognition.