Voiceprint Recognition System Upgrade Without Re-registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing voiceprint recognition systems require users to re-register their voices with each upgrade of the voiceprint feature extraction model, leading to deteriorated recognition performance and a poor user experience, as the new model cannot match the old user template features without re-registration.
Innovation Solution
An electronic device upgrades the voiceprint recognition system by using a verification voice as a new registration voice to update the user template feature and model, allowing for registration-free upgrades that improve both recognition performance and user experience without requiring users to re-enter a registration voice.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a registration procedure is executed again at each upgrade, then voiceprint recognition performance is maintained, but user experience deteriorates
Solution Approach 1:
The system performs preliminary actions by collecting verification voices during normal verification procedures and storing them in advance. When upgrade is needed, these pre-collected verification voices are directly used to update user templates without requiring users to re-register, thus maintaining recognition performance while avoiding the negative user experience of repeated re-registration
Solution Approach 2:
The system makes the verification voice serve dual purposes: both for verification and for template updating during upgrades. The verification voice automatically collected during normal operation is reused to update user templates when new models are deployed, eliminating the need for separate re-registration actions by users
2Ease of operation
If verification voice is used to update user template, then re-registration is eliminated, but model compatibility must be ensured
Solution Approach 1:
The system introduces an intermediary adaptation layer that handles the compatibility between verification voices and new model requirements. This intermediary process includes voice quality detection, feature extraction adaptation, and template updating mechanisms that ensure seamless compatibility without requiring changes to the user-facing verification process
Solution Approach 2:
The system changes parameters of the verification voice processing to match new model requirements during upgrades. This includes adjusting feature extraction parameters, normalization settings, and template update algorithms to ensure compatibility between old verification voices and new voiceprint models while maintaining operational simplicity
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of this application provide an upgrade method, an upgrade apparatus, and an electronic device. The method includes: An electronic device acquires a first verification voice entered by a user; processes the first verification voice by using a first model stored in the electronic device, to obtain a first voiceprint feature; verifies an identity of the user based on the first voiceprint feature and a first user feature template stored in the electronic device; after the identity of the user is verified, if the electronic device has received a second model, processes the first verification voice by using the second model, to obtain a second voiceprint feature; and updates the first user feature template based on the second voiceprint feature, and updates the first model by using the second model. In embodiments of this application, a verification voice obtained in a verification process is used as a new registration voice to complete upgrade and registration of a voiceprint recognition system, so that upgrade of the voiceprint recognition system can be implemented without user perception, and both voiceprint recognition performance and user experience can be improved.