Adaptive Supplementary Information Presentation in Dialogue Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional dialogue support devices lack the ability to dynamically determine the necessity and timing of presenting supplementary information based on user requirements, often disrupting customer service with unnecessary notifications, and fail to differentiate between translated dialogue and supplementary information.
Innovation Solution
An information presentation method that performs speech recognition, translation, and synthesis to generate dialogue text and speech, determines the existence of supplementary information, and presents it only when necessary, using a processor-based system that considers user-specific usage situations, such as language translation frequency and duration of system use.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If supplementary information is presented continuously to support dialogue, then information completeness is improved, but dialogue smoothness deteriorates due to unnecessary interruptions
Solution Approach 1:
The system dynamically adjusts the presentation of supplementary information based on real-time dialogue context and user needs. Instead of continuous presentation, it activates only when necessary conditions are met, balancing information completeness with dialogue smoothness through adaptive control
Solution Approach 2:
The system incorporates feedback mechanisms to monitor dialogue flow and user interaction patterns. By analyzing usage situations and feedback signals, it determines optimal timing for presenting supplementary information, preventing interruptions while ensuring information availability
2Loss of information
If supplementary information is presented frequently to ensure completeness, then information availability is improved, but user attention is reduced due to excessive notifications
Solution Approach 1:
The system changes presentation parameters based on usage situation analysis, including timing, frequency, and modalities of information delivery. By adjusting these parameters dynamically, it maintains information availability while minimizing disruption to user attention and dialogue flow
3Loss of information
If the system translates all dialogue to ensure accuracy, then translation completeness is improved, but processing time increases
Solution Approach 1:
The system applies partial translation strategies by selectively translating only portions of dialogue or only when supplementary information is likely to be presented. This reduces processing time while maintaining sufficient translation accuracy for the intended functionality
Data Source
AI summary
An information presentation method, a non-transitory recording medium storing thereon a computer program, and an information presentation system relate to speech recognition. A speech recognition unit performs speech recognition on speech pertaining to a dialogue and thereby generates dialogue text, a translation unit translates the dialogue text and thereby generates translated dialogue text, and a speech waveform synthesis unit performs speech synthesis on the translated dialogue text and thereby generates translated dialogue speech. An intention understanding unit then determines whether supplementary information exists, based on the dialogue text. If supplementary information exists, a communication unit transmits the supplementary information and the translated dialogue speech to a terminal to present the existence of the supplementary information to at least one person from among a plurality of people, according to the usage situation of the information presentation system of the at least one person.