Dialog Manager Context Store for Multi-Modal Conversations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional dialog managers in spoken language processing systems do not effectively retain and share contextual information across conversations, leading to inefficient user interactions and poor user experience, as they typically require users to re-provide information already given in previous conversations, especially when switching between different applications or functions.
Innovation Solution
Implementing a context store that retains and shares contextual information from prior conversations, allowing the dialog manager to provide necessary information automatically to subsequent applications without prompting the user, and enabling the sharing of additional information across different modalities and applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a conventional dialog manager follows pre-determined scripts and rules to manage user interactions, then the system structure remains simple and manageable, but the user experience deteriorates because users must re-provide information already given in previous conversations
Solution Approach 1:
The system performs preliminary actions by proactively providing information to the user before being asked. The information retrieval module automatically searches for and retrieves relevant contextual information from previous conversations or external sources, and the information presentation module delivers this information to the user proactively, eliminating the need for users to re-provide information they already supplied.
Solution Approach 2:
An information retrieval module is introduced as an intermediary between the dialog manager and the user/conversations. This module acts as a mediator that automatically searches for, retrieves, and presents relevant contextual information from previous conversations or external sources, bridging the gap between stored information and current conversation needs without requiring complex restructuring of the entire dialog manager.
2Loss of time
If the dialog manager prompts the user for all required information in each conversation, then information accuracy is ensured, but the time and effort required for user interaction increases
Solution Approach 1:
The system implements feedback mechanisms where the information presentation module continuously monitors the current conversation context and automatically retrieves relevant information from previous conversations or external sources. This feedback loop ensures that accurate information is provided to users based on their conversation history and current needs, reducing interaction time while maintaining information accuracy through automated contextual awareness.
Solution Approach 2:
The information retrieval module performs preliminary actions by automatically searching for and retrieving relevant contextual information before the user needs to provide it again. This preliminary retrieval of accurate information from previous conversations or external sources reduces the time users spend repeating information while ensuring the accuracy of the information provided through automated verification against conversation history.
3Adaptability or versatility
If the system provides information in multiple modalities, then user experience and accessibility are improved, but the system complexity and resource requirements increase
Solution Approach 1:
The information presentation module is designed with multi-functionality to present information in multiple modalities including text, speech, and visual displays. This universal module can adapt to different user preferences and accessibility needs by delivering the same contextual information through different channels, enhancing user experience and accessibility without requiring separate specialized systems for each modality.
Data Source
AI summary
Features are disclosed for performing functions in response to user requests based on contextual data regarding prior user requests. Users may engage in conversations with a computing device in order to initiate some function or obtain some information. A dialog manager may manage the conversations and store contextual data regarding one or more of the conversations. Processing and responding to subsequent conversations may benefit from the previously stored contextual data by, e.g., reducing the amount of information that a user must provide if the user has already provided the information in the context of a prior conversation. Additional information associated with performing functions responsive to user requests may be shared among applications, further improving efficiency and enhancing the user experience.