Electronic Device Voice Query Short Clip Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multimedia devices provide unnatural text-based voice responses and include irrelevant content in search results when users query video or audio content, leading to meaningless search outcomes.
Innovation Solution
An electronic device and method that analyze user voice commands to extract relevant keywords, allowing the device to generate and provide short clips from original content, focusing only on the user's query by using Endpoint Detection algorithms to edit audio signals and match keywords with specific content sections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If text-based search results are provided using TTS, then voice response capability is achieved, but the response becomes unnatural and less useful
Solution Approach 1:
The patent extracts only the relevant portions of video/audio content that directly answer the user's voice query, rather than providing complete text-based search results. This extraction principle transforms the response from generic text to targeted content segments, improving naturalness while maintaining voice response capability.
Solution Approach 2:
The patent creates short clip copies of the original video/audio content that contain only the relevant portions. These copied segments are then provided as search results, preserving the natural format of the original content while eliminating irrelevant information.
2Loss of information
If original video or audio content is provided as search result, then complete information is available, but irrelevant parts are included making the search result meaningless
Solution Approach 1:
The system extracts specific time segments from the original content that are relevant to the user's query. By taking out only the necessary portions, the patent maintains information completeness for the relevant parts while removing irrelevant content, making search results meaningful and useful.
Solution Approach 2:
The patent segments the original video/audio content into smaller, relevant clips based on the user's voice query. This segmentation allows the system to provide complete information about the queried topic without including unrelated portions of the original content.
3Loss of information
If keyword-based short clip generation is implemented, then relevant content is provided, but additional processing steps are required
Solution Approach 1:
The patent performs preliminary actions by pre-processing the voice query to extract keywords before content generation. This preliminary keyword extraction simplifies the subsequent content selection process, making the overall system more efficient despite the additional processing step.
Solution Approach 2:
The patent introduces keywords as an intermediary between the user's voice query and the video/audio content. This intermediary facilitates the connection by translating natural language queries into searchable terms, enabling efficient content retrieval and reducing the complexity of direct voice-to-content matching.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
An electronic device is disclosed. The electronic device comprises: a communicator for communicating with a server storing information on a plurality of short clips and storing keywords by the plurality of short clips; an outputter; an inputter; and a processor which, when a voice uttered by a user is received via the inputter, transmits a short clip request signal to the server, on the basis of a keyword included in the received uttered voice and information on content outputted from the outputter, and outputs a short clip via the outputter, on the basis of information on the short clip received from the server in response to the request signal.