Cross-Device Voice Control Using Shared Utterance Context
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices struggle to provide continuous and suitable responses to a user's utterances when the user moves between different devices, as they often fail to understand or respond appropriately to subsequent utterances due to lack of domain and intent information extraction.
Innovation Solution
A system and method for electronic devices to cooperate by exchanging utterance-related information over a network, allowing one device to request and receive necessary information from other connected devices to accurately recognize and respond to continuous utterances across different devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If each electronic device independently processes user utterances, then device operation simplicity is maintained, but continuity of service across different devices is lost
Solution Approach 1:
The patent merges the utterance processing capabilities across multiple electronic devices by establishing a network connection where devices share utterance information. When a user speaks to one device, that device analyzes the utterance and shares the result with other devices in the network, ensuring continuous service without requiring complex individual processing at each device.
Solution Approach 2:
The patent implements universality by enabling any electronic device in the network to handle any user utterance type. Each device is configured to process various domains and intents beyond its primary function, allowing seamless transitions between devices while maintaining diverse service capabilities across the network.
2Measurement precision
If an electronic device processes all utterance types locally, then response accuracy is improved, but processing time increases when moving between devices
Solution Approach 1:
The patent applies preliminary action by having the first electronic device analyze the user's utterance before the user moves to the second device. The analysis results, including domain and intent information, are stored and shared in advance, so when the user speaks to the second device, the processing is already complete or near-complete, significantly reducing response time while maintaining accuracy.
Solution Approach 2:
The patent implements feedback by having the first electronic device share its utterance analysis results with the second device. This feedback mechanism allows the second device to receive pre-processed information about the user's intent and domain, enabling it to respond more quickly and accurately without repeating the entire analysis process.
3Measurement precision
If domain and intent information is extracted at every device, then response accuracy is maintained, but information loss occurs when transitioning between devices
Solution Approach 1:
The patent uses the network communication system as an intermediary to transfer domain and intent information between devices. Instead of each device independently extracting this information (which may lead to inconsistencies), the first device extracts the information and transmits it through the network intermediary to the second device, preserving the original context and avoiding information loss during transitions.
Solution Approach 2:
The patent applies preliminary action by having the first electronic device extract and store domain and intent information before the user moves to the second device. This pre-extraction ensures that the contextual information is captured accurately and made available to subsequent devices, preventing information loss that would occur if extraction were delayed or repeated at each device.
Data Source
AI summary
An example controlling method includes receiving an input utterance, determining whether domain information and intent information are able to be extracted by analyzing the input utterance, based on at least one of the domain information and the intent information not being extracted, broadcasting a signal requesting previous utterance related information to one or more external devices connected to a same network as the electronic device, receiving the previous utterance related information from the at least one external device, extracting the domain information and the intent information based on the received previous utterance related information and the input utterance, and obtaining and outputting a response result based on the extracted domain information and intent information.