Voice-Controlled Navigation Device Keyword Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current navigation devices, such as GPS systems, require manual input of destinations, which is inconvenient and potentially dangerous when used while driving, especially with the need to match entire voice commands to digitized samples, leading to time-consuming processes.
Innovation Solution
A voice-controlled navigation device with a keyword database, place name database, voice recognition unit, and electronic map that divides voice commands into segments using keywords and matches them with place names in a search tree to generate navigation routes, allowing for efficient voice-operated navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the whole voice command is matched with each digitized voice sample, then the voice command can be accurately recognized, but the processing time increases significantly
Solution Approach 1:
The patent segments the voice command into multiple keywords based on punctuation marks (commas, periods, question marks). Instead of matching the entire voice command against all digitized samples, the system divides it into smaller keyword units that can be processed independently and efficiently matched against the database.
Solution Approach 2:
The patent extracts keywords from the voice command by identifying punctuation marks as delimiters. This extraction process isolates the essential information-bearing elements (keywords) from the complete voice command, allowing the system to work with only the critical components rather than the entire input string.
2Manufacturing precision
If manual input is used to locate destination on electronic map, then the navigation device can plan route accurately, but the operation becomes troublesome and dangerous when driving
Solution Approach 1:
The patent replaces the mechanical manual input system (typing or pointing on a map interface) with an acoustic input system (voice commands). The voice recognition system converts spoken words into system commands that automatically locate the destination on the electronic map and initiate route planning, eliminating the need for manual interaction with the device interface.
Solution Approach 2:
The system processes the voice command automatically once the keywords are extracted. The navigation device self-services by autonomously interpreting the keywords, locating the destination on the electronic map, and generating the route without requiring further user intervention or manual adjustment.
Data Source
AI summary
In a voice-controlled navigation device and method, a voice command is received, and divided into voice segments Vi (i=1˜n) by comparing with one or more keywords. A voice segment Vi (i=1˜n) is obtained in sequence to be compared with tree nodes in a search tree of place names. A weight value of each tree node is computed according to a comparison, to select one or more tree nodes whose weight values are greater than a predetermined value. Routes formed by all the selected tree nodes are obtained to select a route whose total weight value is the greatest. A navigation to a destination is given by indicating the selected route on an electronic map according to place names represented by the tree nodes of the selected route.