Electronic Device Domain Selection for Speech Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As the number of services increases exponentially, the workload for external servers to identify and process user intents from utterances becomes overwhelming, leading to difficulties in providing continuous speech services when a predetermined domain is unable to process user utterances.
Innovation Solution
An electronic device determines a specific category and domain for processing user utterances, allowing for the selection of an alternative domain if the initial domain is unable to process the utterance, thereby reducing workload and ensuring continuous service delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the external server processes user utterances across all domains, then service coverage is improved, but workload increases exponentially
Solution Approach 1:
The patent segments the domain selection process into two stages: first, the electronic device performs preliminary category identification and domain selection based on local category information; second, the external server performs detailed processing only within the selected domain. This segmentation reduces the external server's workload from evaluating all domains to processing a single selected domain, while maintaining comprehensive service coverage.
Solution Approach 2:
The patent implements preliminary domain selection at the electronic device before transmitting the utterance to the external server. The device uses local category information to pre-identify the most relevant domain, performing the filtering action in advance. This preliminary action reduces the external server's processing burden by eliminating the need to evaluate all domains, while ensuring the selected domain is appropriate for the user's intent.
2Measurement precision
If the external server selects domains for each utterance, then service accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs domain selection in advance at the electronic device using local category information, so the external server receives the utterance with a pre-determined domain. This preliminary domain identification maintains service accuracy by ensuring the correct domain is selected, while reducing processing time by eliminating the external server's need to evaluate multiple domains in real-time.
Solution Approach 2:
The electronic device performs self-service by autonomously identifying the category and selecting the appropriate domain using its local category information, without requiring the external server to perform domain selection. This self-service capability at the device level maintains accurate domain selection while significantly reducing the external server's processing time and computational burden.
3Adaptability or versatility
If multiple domains are registered in the external server, then service versatility is improved, but domain identification complexity increases
Solution Approach 1:
The patent segments the domain identification task between the electronic device and the external server. The device handles category identification and domain selection using local category information, while the external server handles service execution within the selected domain. This segmentation allows the external server to maintain a comprehensive registry of multiple domains for versatility, while the device manages the complexity of domain identification locally.
Solution Approach 2:
The patent introduces category information as an intermediary layer between user utterances and domains. The electronic device uses this intermediate category representation to select the appropriate domain, simplifying the domain identification process. The external server maintains the comprehensive domain registry but only processes within the selected domain, reducing its identification complexity while preserving service versatility.
4Speed
If the electronic device identifies domains locally, then processing speed is improved, but accuracy decreases
Solution Approach 1:
The patent performs preliminary domain selection at the electronic device using local category information, achieving fast processing speed. The device leverages pre-stored category information to quickly identify the relevant domain without requiring real-time analysis of all domains. This preliminary action maintains accuracy by using the category information as a reliable basis for domain selection, while achieving fast processing at the device level.
Data Source
AI summary
According to various embodiments, provided is a control operation comprising: receiving a first utterance from a device; determining a specific category corresponding to the received first utterance from among a plurality of categories; determining a first domain for processing of the first utterance from among a plurality of domains related to the determined specific category; acquiring information enabling an operation corresponding to the first utterance to be performed based on the determined first domain; receiving a second utterance from the device; identifying whether processing of the second utterance is possible in the determined first domain; based on identifying that the processing of the second user utterance based on the first domain is not possible, checking a second domain among the plurality of domains, and identifying whether processing of the second utterance based on the second domain is possible; and based on identifying that the processing of the second utterance based on the second domain is possible, acquiring information enabling an operation corresponding to the second utterance to be performed based on the second domain. An operation of determining a domain corresponding to a user utterance of the electronic device may be performed using an artificial intelligence model.