Shared Audio Models for Distributed User Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Excessive network transmissions of network traffic data can lead to inefficient bandwidth utilization and degrade the quality of response due to processing capacity limitations, complicating data routing and preventing timely processing of network traffic data.
Innovation Solution
A data processing system that shares audio identification models among client computing devices, enabling accurate user identification and reducing computational and network resource usage by generating models once and updating them with audio samples from different locations, thus optimizing resource utilization and improving identification accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If each client computing device generates and processes its own audio identification models independently, then user identification can be performed locally without network dependency, but network traffic increases significantly and processing capacity is wasted due to redundant model generation and updates across multiple devices
Solution Approach 1:
The patent merges the audio identification model generation and update functions into a centralized server that serves multiple client devices. The server consolidates audio samples from various clients, generates a single unified audio identification model, and distributes it to all clients. This eliminates redundant model generation at each client device while maintaining local identification capability through model distribution, thereby reducing network traffic and processing waste.
Solution Approach 2:
The server implements a universal audio identification model that serves multiple client computing devices simultaneously. Instead of each client having its own separate model generation capability, the server provides a multi-functional platform that handles model generation, updates, and distribution for all clients. This universal approach reduces overall system resource consumption while maintaining identification reliability across all devices.
2Measurement precision
If audio identification models are updated frequently across multiple client devices, then user identification accuracy improves, but network traffic increases and processing capacity is exceeded
Solution Approach 1:
The patent consolidates the model update process into a centralized server that collects audio samples from multiple clients, processes them collectively to generate updated audio identification models, and distributes the updates. This merging approach ensures that processing capacity is utilized efficiently at the server level rather than being duplicated across multiple client devices, allowing frequent updates without exceeding individual device processing limits while maintaining identification accuracy.
3Ease of operation
If multiple client devices each maintain their own audio identification models, then local processing can be performed without network dependency, but redundant computations increase and resource utilization becomes inefficient
Solution Approach 1:
The patent segments the audio identification system into two distinct functional parts: model generation/update operations centralized at the server, and model execution/distribution operations at client devices. This segmentation allows clients to maintain simple local processing capabilities by receiving pre-generated models from the server, while complex model generation and update computations are performed centrally. The result is efficient resource utilization with clients performing only lightweight local identification operations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure is generally directed to a data processing system for customizing content in a voice activated computer network environment. With user consent, the data processing system can improve the efficiency and effectiveness of auditory data packet transmission over one or more computer networks by, for example, increasing the accuracy of the voice identification process used in the generation of customized content. The present solution can make accurate identifications while generating fewer audio identification models, which are computationally intensive to generate.