Voice User Identification via Amalgamated Probability Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice-controllable devices face challenges in accurately identifying and authenticating multiple registered users, especially when used by guests or in environments where unauthorized users may inadvertently interact with the system, due to the short duration of registration and in-use voice samples.
Innovation Solution
A method and system that utilize a Machine Learning Algorithm (MLA) to analyze voice features and user frequency analysis to generate amalgamated probability parameters, selecting the most likely registered user based on these probabilities, and managing access privileges accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If voice-controllable devices use short duration voice samples for registration and authentication, then the registration process is quick and user-friendly, but the accuracy of user identification deteriorates
Solution Approach 1:
The patent combines multiple probability parameters (first probability from voice feature analysis, second probability from user frequency analysis, and third probability from contextual analysis) into an amalgamated probability value. This merging of multiple assessment dimensions compensates for the limited information available in short voice samples, enabling accurate user identification without requiring longer registration times.
Solution Approach 2:
The patent introduces probability parameters as intermediary metrics that bridge the gap between short voice samples and reliable user identification. Instead of directly comparing voice samples, the system uses probability values derived from multiple analysis methods to mediate the identification process, improving accuracy while maintaining quick registration.
2Adaptability or versatility
If the system authenticates multiple registered users with different access privileges, then the system becomes more versatile and user-specific, but the complexity of user management increases
Solution Approach 1:
The patent changes the state of user authentication from binary (authenticated/not authenticated) to a probabilistic scale (amalgamated probability value). This parameter transformation enables nuanced user identification and facilitates differentiated access privilege management, allowing the system to provide user-specific customization while managing multiple users through a unified probabilistic framework.
3Ease of operation
If the system is accessible to guests and passersby, then the ease of operation improves, but the risk of unauthorized access increases
Solution Approach 1:
The patent applies preliminary anti-action by establishing probability thresholds before guest interactions can result in unauthorized access. The system pre-defines acceptable probability ranges that distinguish legitimate guests from unauthorized users, preventing harmful actions before they occur. This approach maintains ease of operation for genuine guests while blocking unauthorized access attempts.
Solution Approach 2:
The system uses feedback mechanisms where the amalgamated probability values provide continuous assessment of user authenticity. This feedback loop allows the system to dynamically adjust access decisions based on real-time analysis, maintaining accessibility for legitimate users while preventing unauthorized access through probabilistic validation at each interaction point.
Data Source
AI summary
There are disclosed methods and systems for determining a speaker of a set of registered users associated with a voice-controllable device. The method is executable by an electronic device configured to execute a Machine Learning Algorithm (MLA). The method comprises executing the MLA to determine a first probability parameter indicative of the speaker of the user utterance being one of the set of registered users; executing a user frequency analysis to generate, for each given one of the set of registered users, a second probability parameter the being an apriori frequency based probability; generating, for the electronic device, for each given one of the set of registered users an amalgamated probability based on the first probability and the second probability associated therewith; selecting the given one of the set of registered users as the speaker of the user utterance based on the amalgamated probability value.