Multi-Speechlet Response Context Transfer via Navigator Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current speech recognition systems are limited in their ability to leverage multiple speechlets to respond to a single utterance or within a single session, leading to fragmented user experiences and inefficient data handling.
Innovation Solution
The system enables multiple speechlets to process user inputs within a single session by using navigator objects to maintain context and pass user experiences between speechlets, allowing one speechlet to provide data for processing by another, while ensuring secure handling of sensitive information through sandboxed components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single speechlet processes all user inputs within a session, then the system architecture remains simple, but the system cannot leverage multiple specialized speechlets and loses contextual efficiency
Solution Approach 1:
The system segments the session management process by introducing navigator objects that independently track context for each speechlet invocation. This allows multiple specialized speechlets to be invoked within a single session without requiring complex centralized coordination, as each navigator object maintains its own context stack independently.
Solution Approach 2:
Navigator objects serve as intermediaries between the session manager and individual speechlets. These navigator objects handle context passing and state management, allowing speechlets to interact with the session without direct complex communication between them, thus reducing overall system complexity while enabling multi-speechlet functionality.
2Loss of information
If multiple speechlets process user inputs within a single session, then data sharing and user experience continuity improve, but context management and data security become more complex
Solution Approach 1:
The system implements nested context management where navigator objects contain stacks of context data that are nested within the session structure. Each speechlet invocation creates a nested context layer that can access and modify parent context levels, enabling efficient context retention without flat complex data structures.
Solution Approach 2:
Each navigator object maintains local context quality tailored to specific speechlet requirements. Context data is organized with local quality attributes that allow different speechlets to access only the context relevant to their function, reducing unnecessary data processing and management complexity while improving context retention efficiency.
3Adaptability or versatility
If speechlets have broad access to session data, then processing flexibility increases, but security and data protection decrease
Solution Approach 1:
The system applies local quality control to data access by associating specific data permissions with each navigator object and speechlet invocation. Speechlets access only the context data locally relevant to their processing needs, maintaining security while preserving processing flexibility for authorized operations.
Solution Approach 2:
Session data is segmented into context stacks that are independently managed by each navigator object. This segmentation isolates sensitive data within specific context layers, allowing flexible processing within authorized boundaries while preventing unauthorized access to segmented data portions.
Data Source
AI summary
Techniques for transferring control of a system-user dialog session are described. A first speechlet component may interact with a user until the first speechlet component receives user input that the first speechlet component cannot handle. The first speechlet component may output an action representing the user input. A system may determine a second speechlet component configured to execute the action. The system may send the second speechlet component a navigator object that results in the second speechlet component handling the user interaction that the first speechlet component could not handle. Once the second speechlet component is finished processing, the second speechlet component may output an updated navigator object, which causes the first speechlet component to either further interact with a user or cause a current dialog session to be closed. The system may additionally maintain a data structure representing calling speechlet components and called speechlet components associated with the session.