Speech Guidance Device Visual Recognition Difference Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing route guidance devices struggle to effectively inform drivers when their visual recognition object differs from the intended guidance object, leading to difficulty in recognizing the correct branching road during navigation.
Innovation Solution
A speech guidance device that utilizes a processor and memory to detect the driver's visual recognition object, compare its attributes with the intended guidance object, and generate a guidance sentence that emphasizes the differing attributes, with the option to adjust the sentence based on word familiarity to enhance recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If normal reading speech is used to guide the driver's visual line to the branching road, then the speech guidance can be provided, but it is difficult for the driver to notice that the visual recognition object is different from the guidance object
Solution Approach 1:
The patent applies acoustic equivalent of 'color changes' by modifying the speech output to emphasize specific words. When the driver's visual recognition object differs from the guidance object, the system highlights the differing attribute words in the speech guidance (e.g., emphasizing 'bank' vs. 'hillside' in landmark descriptions), making the difference noticeable to the driver through prosodic changes in the speech signal.
Solution Approach 2:
The system changes parameters of the speech output by detecting differences between visual recognition objects and guidance objects, then modifying the speech synthesis parameters to emphasize the differing attributes. This involves analyzing object attributes (such as landmark types, road names) and adjusting speech emphasis parameters accordingly to highlight discrepancies.
2Loss of information
If the speech guidance emphasizes the differing attribute, then the driver can easily recognize the difference, but the speech processing complexity increases
Solution Approach 1:
The system applies partial action by selectively emphasizing only the specific words or phrases that represent differing attributes between the visual recognition object and the guidance object, rather than emphasizing the entire speech guidance. This targeted approach reduces unnecessary processing complexity while still achieving effective difference recognition.
Solution Approach 2:
The patent introduces an intermediary processing layer that compares object attributes and determines which words require emphasis. This intermediary module acts as a mediator between the guidance system and the speech output, selectively identifying differing attributes and translating them into emphasis instructions for the speech synthesis system.
3Ease of operation
If the guidance sentence is adjusted based on word familiarity, then the guidance becomes more intuitive, but the processing time increases
Solution Approach 1:
The system performs preliminary action by pre-storing familiarity information for multiple words used in guidance sentences. This pre-computed familiarity data is readily available when generating speech guidance, allowing the system to quickly adjust sentence structure based on word familiarity without performing complex real-time analysis, thus reducing processing time.
Data Source
AI summary
A speech guidance device includes; a guidance object determining unit for outputting a guidance object to which a visual line of a user is guided; a visual recognition object detecting unit for detecting a visual recognition object which the user is viewing from information indicating a direction of the visual line of the user; a difference detecting unit for detecting a difference between an attribute of the visual recognition object and an attribute of the guidance object; a generating unit for generating a guidance sentence including the attribute of the guidance object; and a synthesizing unit for, when there is a difference between the attribute of the guidance object included in the guidance sentence and the attribute of the visual recognition object, synthesizing and outputting speech of the guidance sentence emphasizing a word of the attribute having the difference.