Dialog-Based Enrollment for Automated Assistant User Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated assistants require manual graphical user interface interaction for enrollment and rely on voice matching technology, which is not reliable in noisy environments or with multiple speakers, and often require default hot words for invocation, limiting their ability to distinguish between users and provide secure access to sensitive features.
Innovation Solution
Implement a dialog-based enrollment system that uses human-to-computer dialog to enroll users through visual and voice profiles, determining trust levels based on sensor data and historical interaction, allowing dynamic hot words and feature access based on recognition confidence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual graphical user interface interaction is used for enrollment, then enrollment accuracy is improved, but ease of operation deteriorates
Solution Approach 1:
The patent replaces manual graphical user interface interaction with automated dialog-based enrollment. The system uses speech recognition and natural language processing to automatically identify users, capture their voice profiles, and enroll them without requiring manual configuration. This substitution of mechanical manual operations with automated dialog-based processes resolves the contradiction by maintaining accurate enrollment while dramatically improving ease of operation.
Solution Approach 2:
The patent implements self-service enrollment where users automatically enroll themselves through natural language dialog. The system detects when a user interacts with the automated assistant, automatically initiates the enrollment process, captures voice samples, and stores profiles without requiring the user to manually navigate configuration interfaces. This self-service approach resolves the contradiction by enabling accurate enrollment while eliminating the need for manual operation.
2Ease of operation
If voice matching technology is used to distinguish between individuals, then ease of operation is improved, but reliability deteriorates
Solution Approach 1:
The patent merges multiple recognition modalities - voice matching, facial recognition, and contextual analysis - into a unified enrollment and recognition system. By combining these different recognition approaches, the system achieves both ease of operation (through automatic multi-modal recognition) and improved reliability (through cross-validation and complementary strengths of different modalities).
Solution Approach 2:
The patent changes the parameters of user identification from relying solely on voice characteristics to incorporating multiple parameters including facial features, contextual information, and interaction patterns. This multi-parameter approach improves reliability by not depending on a single potentially unreliable voice match, while maintaining ease of operation through automated processing of multiple parameters simultaneously.
3Device complexity
If default hot words are used for invocation, then device complexity is reduced, but adaptability deteriorates
Solution Approach 1:
The patent implements dynamic hot words that are automatically customized for each user based on their voice profile and preferences. Instead of using a fixed set of default hot words, the system learns and adapts the invocation phrases to match each user's speaking patterns and preferences. This dynamic adaptation resolves the contradiction by maintaining relatively simple device architecture while significantly improving adaptability to individual users.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system monitors and learns from user interactions to automatically update and refine the hot word list for each user. Through continuous feedback from actual usage patterns, the system adapts the invocation phrases to better match each user's preferences, resolving the contradiction between device simplicity and user-specific adaptability.
4Adaptability or versatility
If multiple enrollment methods are implemented, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent implements a universal enrollment system that handles multiple enrollment methods (voice-based, facial recognition, contextual) through a single unified framework. The same core enrollment architecture accommodates different modalities by processing various input types through common algorithms and data structures, thereby achieving high adaptability without proportionally increasing device complexity.
Solution Approach 2:
The patent introduces an intermediary layer of automated dialog-based processing that mediates between different enrollment modalities and the core recognition system. This intermediary layer standardizes the interface for multiple enrollment methods while managing their complexity internally, allowing the system to support diverse enrollment approaches without exposing the increased complexity to the user or requiring proportional increases in overall device complexity.
Data Source
AI summary
Techniques are described herein for dialog-based enrollment of individual users for single-and/or multi-modal recognition by an automated assistant, as well as determining how to respond to a particular user's request based on the particular user being enrolled and/or recognized. Rather than requiring operation of a graphical user interface for individual enrollment, dialog-based enrollment enables users to enroll themselves (or others) by way of a human-to-computer dialog with the automated assistant.