Voice Command Application Ranking via Bidding Weights and Power Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition systems face challenges in accurately identifying action keywords within voice commands and executing the most appropriate application without requiring additional user interaction, as they lack efficient methods to rank and prioritize candidate applications based on relevance and user feedback.
Innovation Solution
The system receives audio data, determines action and object keywords, and ranks candidate applications using a combination of bidding weights, power scores, and feedback scores to identify and execute the highest-ranked application associated with the keywords, ensuring the most relevant application is launched for the user's command.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system ranks candidate applications using multiple factors (bidding weight, power score, feedback score), then the accuracy of selecting the most appropriate application is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent segments the application selection process into distinct components: bidding weight calculation, power score determination, and feedback score evaluation. Each component handles a specific aspect of application ranking independently, allowing the system to maintain high selection accuracy while managing computational complexity through modular processing of each scoring factor
Solution Approach 2:
The system performs preliminary actions by pre-calculating and storing bidding weights, power scores, and feedback scores for candidate applications before the actual voice command execution. This allows the ranking process to efficiently retrieve and combine pre-computed values rather than calculating everything in real-time, reducing processing complexity while maintaining selection accuracy
2Reliability
If the system requires additional user interaction to confirm application selection, then the reliability of execution is improved, but the user experience and operational efficiency deteriorate
Solution Approach 1:
The patent implements feedback mechanisms where the system monitors user actions and app launches to update feedback scores for URIs. This continuous feedback loop allows the system to learn from user behavior patterns and improve its application selection accuracy over time, enabling confident automatic execution without requiring explicit user confirmation for each command while maintaining high reliability
Solution Approach 2:
The system performs self-service by automatically selecting and executing the highest-ranked application based on its multi-factor ranking algorithm, without requiring user confirmation. The system serves itself by making intelligent decisions based on accumulated data from bidding weights, power scores, and feedback scores, thereby improving ease of operation while maintaining reliability through data-driven selection
3Ease of operation
If the system considers only popular applications with high ratings, then the ease of operation is improved, but the adaptability to specific user needs and contexts deteriorates
Solution Approach 1:
The patent applies local quality by making the bidding weight application-specific and context-dependent. Each application can have customized bidding weights for different URIs based on its specific functionality and the user's interaction patterns. This allows the system to maintain simplicity in the overall ranking process while adapting to specific user needs and application contexts through localized weight adjustments
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving audio data corresponding to an utterance of a user, determining that at least a portion of the audio data corresponds to an action keyword, obtaining candidate applications that correspond to the action keyword, which include applications associated with a uniform resource identifier (URI) that corresponds to the action keyword, ranking the candidate applications that correspond to the action keyword based at least on: (i) a bidding weight associated with the action keyword for each of the candidate applications, and (ii) a power score associated with each of the candidate applications, identifying a highest ranked application among the candidate applications that has a URI that corresponds to at least the action keyword, and executing the URI for the highest ranked application among the candidate applications that correspond to at least the action keyword.