Smart TV Server Query Response Diversity via Dynamic Text Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current smart devices, such as smart televisions, respond to user queries with limited and manually preset templates, resulting in uninteresting and inflexible responses that fail to stimulate user interest in audio-visual content, leading to a suboptimal user experience.
Innovation Solution
A smart television system that sends query commands to a server, which determines media resource information and generates reply texts using attribute information and preset templates, ensuring that consecutive queries for the same content receive different responses, enhancing user engagement through varied and dynamic content presentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manually preset templates are used for responding to user queries, then the system structure is simple and easy to implement, but the response diversity is limited and user experience is poor
Solution Approach 1:
The patent applies dynamics by making the reply text generation process variable and adaptive. Instead of using fixed manually preset templates, the system dynamically generates diverse reply texts by combining extracted attribute information with a pool of template sentences. The controller randomly selects and combines these elements to produce different reply texts for the same query, thereby increasing response diversity while maintaining system manageability through the use of structured attribute extraction and template-based generation.
2Ease of operation
If fixed reply texts are used for the same audio/video content, then the system is easy to control and maintain, but user interest decreases due to repetitive responses
Solution Approach 1:
The system implements dynamics by enabling the same audio/video content to generate multiple different reply texts. The controller extracts attribute information from the query and combines it with randomly selected template sentences from a pool, ensuring that repeated queries for the same content produce varied responses. This maintains system controllability through structured attribute extraction and template management while significantly improving response variety and user engagement.
Solution Approach 2:
The patent applies parameter changes by varying the reply text parameters (such as sentence structure, wording, and expression) while keeping the core information consistent. The system changes the textual parameters by selecting different template sentences and combining them with extracted attributes in different ways, thereby producing diverse responses without altering the fundamental meaning or control structure of the system.
3Productivity
If simple query-response system is used, then the implementation is straightforward, but the user interaction quality is insufficient
Solution Approach 1:
The patent applies segmentation by dividing the reply text generation process into distinct modules: attribute extraction module, template selection module, and text combination module. The controller segments the query processing into extracting specific attribute information (such as audio/video title, type, and other relevant attributes) and then combining these attributes with appropriate template sentences. This modular segmentation improves interaction quality through diverse responses while keeping implementation manageable through clear module separation.
Data Source
AI summary
The present disclosure discloses a smart television and a server. In the present disclosure, a query command comprising a keyword is sent to a server in response to a query request for an audio/video from the user input interface, wherein the server is configured to determine media resource information of corresponding target audio/video according to the keyword in response to receiving the query command, and generate a reply text comprising comment information of the target audio/video according to the keyword; the media resource information and the reply text sent from the server are received; and the display is controlled to present the media resource information in a first area on the user interface and present the reply text in a second area on the user interface.