Starting-Word Speaker Recognition for Personalized Display Results
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speaker recognition systems do not determine whether to recognize a speaker based on the recognition of a preset starting word, leading to inconsistent display of personal information and search results.
Innovation Solution
A display device equipped with a network interface and controller that communicates with an NLP server to identify a starting word, allowing it to provide personalized search results linked to the speaker's account information when a specific starting word is recognized.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If speaker recognition is always performed regardless of starting word type, then speaker identification capability is improved, but system complexity and processing overhead increase
Solution Approach 1:
The system dynamically adjusts speaker recognition behavior based on the type of starting word detected. For first type starting words, speaker recognition is activated to provide personalized results. For second type starting words, speaker recognition is skipped to reduce complexity. This dynamic adaptation resolves the contradiction by making the system flexible rather than static.
Solution Approach 2:
The system changes the operational parameter of speaker recognition (on/off state) based on the starting word type. This parameter change allows the system to optimize between reliability and complexity by enabling speaker recognition only when necessary (for first type starting words) and disabling it when not needed (for second type starting words).
2Ease of operation
If personalized search results are provided for all voice inputs, then user experience is improved, but processing time and resource consumption increase
Solution Approach 1:
The system applies partial action by providing personalized search results only for a subset of voice inputs (those starting with first type starting words). For other inputs (second type starting words), the system uses standard processing without personalized results. This partial application of personalization resolves the contradiction by delivering user experience improvements only where necessary while avoiding unnecessary processing overhead.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A display device according to an embodiment of the present disclosure may include a display; a network interface configured to communicate with a Natural Language Processing (NLP) server; and a controller configured to obtain first voice data uttered by a first speaker, receive a first response result linking with account information of the first speaker from the NLP server based on the fact that the first voice data includes a first type of starting word, and display the first response result on the display.