Voiceprint Authentication Using Dynamic Text Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voiceprint recognition-based identity authentication methods are vulnerable to security hazards as they require pre-set text information to be disclosed, allowing potential unauthorized access by recording and playing back voice files.

Innovation Solution

A method and system that acquire historical voice files from user interactions, perform filtering and text recognition to generate reference voiceprint information, which includes both voice and text data, stored with a unique identifier, allowing for secure and accurate authentication by comparing unknown voice information against randomly selected reference voiceprint data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If pre-set text information is disclosed for voiceprint authentication, then the authentication process can be completed, but security vulnerabilities arise allowing unauthorized access through voice file recording and playback

Engineering Contradiction:
Improveauthentication securityVSAvoidtext information disclosure
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system performs text recognition on historical voice files during the voiceprint library establishment phase, before authentication occurs. This preliminary extraction of text information from actual user speech eliminates the need to pre-disclose authentication text, as the text is dynamically obtained from recorded conversations rather than being predetermined and exposed to users

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Instead of using pre-set text information that could be recorded and replayed, the system captures and processes actual voice data from historical communications. The text information is copied from genuine user speech patterns in real conversations, creating authentic voiceprint references that cannot be replicated by simple voice file playback attacks

Inventive Principle:
Principle #26Copying

2Reliability

If historical voice files are processed to extract text information dynamically, then authentication security is improved, but system complexity increases due to additional processing steps

Engineering Contradiction:
Improveauthentication securityVSAvoidprocessing system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system utilizes existing historical voice files from user communications that are already stored in the platform. By processing these pre-existing files through text recognition, the system generates authentication data without requiring separate recording sessions or additional user interactions, making the security enhancement self-sufficient and eliminating the need for complex additional hardware or user-facing components

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS10593334B2Method and apparatus for generating voiceprint information comprised of reference pieces each used for authentication
Publication Date: 2020.03.17 ADVANCED NEW TECHNOLOGIES CO LTD
  • US10593334B2 patent drawing
  • US10593334B2 patent drawing
  • US10593334B2 patent drawing

AI summary

A method for generating voiceprint information is provided. The method includes acquiring a historical voice file generated by a call between a first user and a second user; executing text recognition processing on the voice information to obtain text information corresponding to the voice information; and storing the voice information and the corresponding text information as reference voiceprint information of the first user, and storing an identifier of the first user. Furthermore each voiceprint information comprises a plurality of pieces of reference voiceprint information, each of which is sufficient to authenticate a user.