Display Voice Recognition With Multi-Source Candidate Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice recognition systems in electronic devices have limitations such as fixed input means, slow processing speed due to large candidate instruction word sets, and limited storage capacity, making them inefficient for various input methods and recognition rates.
Innovation Solution
A display apparatus with multiple voice acquiring units, including built-in and external microphones, communicates with external devices to receive candidate instruction words, enabling dynamic management and recognition of user voices through a centralized voice recognition engine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple voice acquiring units are provided to enable various input means, then adaptability is improved, but device complexity increases
Solution Approach 1:
The voice recognition system is segmented into multiple independent voice acquiring units (built-in microphone, external microphone, mobile phone) that can be selectively activated. Each unit operates independently but contributes to the same recognition goal, allowing the system to handle various input methods without requiring a completely integrated complex structure.
Solution Approach 2:
The controller is designed to universally handle voice inputs from multiple different sources (built-in microphone, external microphone, mobile phone). The same voice recognition engine processes commands regardless of which acquiring unit detected the voice, making the system multi-functional without proportionally increasing complexity.
2Measurement precision
If many candidate instruction words are provided to increase recognition rate, then recognition accuracy is improved, but processing speed deteriorates
Solution Approach 1:
Candidate instruction words are pre-stored in the external device's memory before the voice recognition process begins. When a voice input is detected, the system quickly retrieves and compares against these pre-prepared candidates rather than generating or searching for them during processing, significantly reducing processing time while maintaining high recognition accuracy.
3Measurement precision
If candidate instruction words are stored in the main apparatus to improve recognition, then recognition accuracy is improved, but storage capacity requirements increase
Solution Approach 1:
The storage function for candidate instruction words is extracted from the main apparatus (display device) and relocated to external devices (mobile phone, external microphone device). The main apparatus only needs to temporarily hold received voice signals and comparison results, while the bulk storage responsibility is taken out and assigned to external devices with larger storage capacity.
Solution Approach 2:
The controller acts as an intermediary between the voice acquiring units and the storage/retrieval operations. It receives voice inputs, requests candidate instruction words from external devices, manages the comparison process, and coordinates the flow of information without requiring the main apparatus to permanently store large datasets.
Data Source
AI summary
Disclosed are a display apparatus, a voice acquiring apparatus and a voice recognition method thereof, the display apparatus including: a display unit which displays an image; a communication unit which communicates with a plurality of external apparatuses; and a controller which includes a voice recognition engine to recognize a user's voice, receives a voice signal from a voice acquiring unit, and controls the communication unit to receive candidate instruction words from at least one of the plurality of external apparatuses to recognize the received voice signal.