AI Audio Identification With Distributed Verification Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio identification techniques face challenges in securing reliability and objectivity of audio information due to personal variations among employees, limited range of data collection, and regional limitations, leading to decreased performance.
Innovation Solution
An audio data identification apparatus that randomly collects audio data through networks, uses AI algorithms to parse and match identification information, and receives feedback from unspecified users to train the algorithm, improving reliability and range of identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If audio information is collected by the method of collecting answers from employees, then the collection process is simple to implement, but the reliability and objectivity of the collected audio information deteriorates due to personal characteristics and limited range
Solution Approach 1:
The system enables unspecified users to automatically verify audio information through their terminals without requiring manual verification by employees. The verification is performed autonomously by the users' devices, eliminating the need for human intervention in the verification process while maintaining high reliability through distributed verification
Solution Approach 2:
The system creates copies of audio information and distributes them to multiple unspecified users for verification. Each user receives and verifies copies of the same audio data, allowing parallel verification across multiple independent terminals, which improves both reliability and efficiency
2Ease of manufacture
If audio information is collected by the method of collecting answers from employees, then the implementation process is straightforward, but the range and diversity of collected audio information deteriorates
Solution Approach 1:
The system is designed to work with unspecified users rather than requiring specific employees, making it universally applicable to any user base. This multi-functionality allows the system to collect audio information from diverse sources and contexts, significantly expanding the range and diversity of collected data
Solution Approach 2:
The system transitions from a single-dimension approach (employee-based collection) to a multi-dimensional approach by distributing verification across multiple unspecified users. This dimensional expansion allows simultaneous collection from various user contexts, regions, and scenarios, greatly enhancing the diversity and range of audio information
3Ease of manufacture
If the method of collecting answers is performed passively by employees, then the implementation is simple, but the productivity of collecting audio information deteriorates due to considerable time required
Solution Approach 1:
The system implements periodic verification where unspecified users automatically verify audio information at scheduled intervals or triggered by specific conditions. This periodic automated verification replaces continuous passive employee monitoring, significantly improving productivity while maintaining simplicity
Solution Approach 2:
The system enables continuous automated verification by unspecified users without interruption, replacing the discontinuous and time-consuming passive employee verification process. The useful action of verification continues uninterrupted across multiple users and time periods, dramatically increasing productivity
4Device complexity
If employees collect audio information in a limited number, then the collection process is manageable, but the performance of audio identification technique deteriorates due to insufficient training data for AI algorithm
Solution Approach 1:
The verification process is segmented and distributed across multiple unspecified users rather than being集中 in a small number of employees. Each user performs a portion of the verification task, allowing the system to handle large volumes of audio information while maintaining manageable complexity through distributed processing
Solution Approach 2:
The system transitions from a limited employee-based collection to a multi-dimensional distributed verification model involving unspecified users. This dimensional expansion enables the system to process and verify large quantities of audio data across multiple users and contexts, improving AI training data quality without proportionally increasing process complexity
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The present invention relates to an audio data identification apparatus for collecting random audio data and identifying an audio resource obtained by extracting any one section of the collected audio data. The audio data identification apparatus comprises: a communication unit that collects and transmits the random audio data; and a control unit that identifies the collected audio data. The control unit comprises: a parsing unit that parses the collected audio data into predetermined units; an extraction unit that selects, as the audio resource, any one of a plurality of parsed sections of the audio data; a matching unit that matches identification information of the audio resource via a pre-loaded artificial intelligence algorithm; and a verification unit that verifies the identification information matched to the audio resource.