Dialog Frame Engine for Context-Aware Voice Task Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mobile devices struggle to accurately interpret and execute user voice commands for tasks like creating calendar events or searching for information, often requiring multiple clarifying questions and lacking context understanding.
Innovation Solution
The system employs a dialog frame engine that generates frames for tasks, processes user speech to identify necessary values, and uses search engines to provide context-aware responses, allowing users to answer questions with related information that satisfies task requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional speech recognition software is used to interpret user commands, then the system can execute basic tasks like calling or searching, but the system lacks context understanding and requires multiple clarifying questions
Solution Approach 1:
The system performs preliminary actions by generating a task frame with expected value types before processing user responses. This frame structure is prepared in advance based on the task type, allowing the system to anticipate what information is needed and how to interpret user answers in context, thereby reducing the need for multiple clarifying questions
Solution Approach 2:
The system uses feedback by comparing user speech against the pre-defined task frame structure. The frame specifies expected value types for various parameters, and the system continuously checks whether user responses satisfy these expectations, allowing for context-aware interpretation and reducing information loss during the interaction
2Measurement precision
If the system asks for clarifying information to complete instructions, then task accuracy improves, but user interaction complexity increases
Solution Approach 1:
The system segments the task completion process into distinct value types (e.g., location, time, date) with each having its own frame structure and validation rules. This segmentation allows the system to handle each piece of information independently with specific validation, improving accuracy without requiring the user to understand the overall complex interaction structure
Solution Approach 2:
The task frame structure serves multiple functions: it defines expected input formats, validates user responses, provides context for interpretation, and guides the dialogue flow. This universal structure handles diverse task types (calls, texts, searches, reminders) through a common mechanism, reducing interaction complexity while maintaining precision
3Speed
If the system processes each speech request independently, then processing speed is maintained, but context from previous statements is lost
Solution Approach 1:
The system performs preliminary action by establishing the task frame structure before processing individual user responses. This pre-defined structure contains all necessary context about what information is needed and in what format, allowing rapid processing of each statement while maintaining context through the persistent frame structure that carries information across multiple interactions
Solution Approach 2:
The task frame structure serves itself by automatically maintaining context across statements. The frame persists and accumulates satisfied and unsatisfied value types, allowing the system to process each user input independently while the frame itself maintains the contextual state, eliminating the need for complex inter-statement analysis
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using frames for performing tasks. One of the methods includes receiving a first request to perform a task, the first request comprising user speech identifying the task; generating a frame associated with the task, wherein the frame comprises one or more types of values necessary to perform the task, and wherein each type of value can be satisfied by a respective value; receiving a second request to provide information related to a question, the second request comprising user speech identifying the question; providing information identifying the question to a search engine, and receiving a response identifying one or more terms; determining that at least one term can satisfy a type of value necessary to perform the task; and storing the at least one term in the frame.