Biometric User Recognition System with Acoustic Transfer Function
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional checkout processes in retail and distribution settings require customers to carry and use physical objects for payment or identification, which can be inconvenient and inefficient, especially in environments where cashiers or dedicated self-checkout stands are not available.
Innovation Solution
A user-recognition system that utilizes facial-recognition and speech-recognition techniques to identify enrolled users, allowing them to check out by looking at a camera and speaking their name, eliminating the need for physical identification methods and enabling automated charging of user accounts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional checkout processes use physical objects for identification and payment, then the checkout process can be completed, but customer convenience and efficiency deteriorate due to the need to carry and use physical items
Solution Approach 1:
The patent replaces the mechanical system of physical identification objects (cards, cash) with an automated optical recognition system using cameras and image processing to capture and identify customers, eliminating the need for physical items during checkout
Solution Approach 2:
The system enables self-service checkout by automatically identifying customers through facial recognition and capturing their information without requiring assistance from cashiers or dedicated checkout stands, allowing customers to checkout independently
2Productivity
If physical stores use cashiers or dedicated self-checkout stands, then transactions can be processed, but device complexity and operational overhead increase
Solution Approach 1:
The patent applies multi-functionality by using a single camera-based system to perform multiple functions including customer identification, payment processing, and account charging, eliminating the need for separate dedicated checkout stands and reducing overall infrastructure complexity
Solution Approach 2:
The system extracts the core identification and transaction processing functions from complex physical checkout infrastructure and implements them through a simplified camera-based automated system, removing unnecessary hardware and operational overhead
Data Source
AI summary
This disclosure describes techniques for identifying users that are enrolled for use of a user-recognition system. To be identified using the user-recognition system, a user may first enroll in the system by stating an utterance at a first device having a first microphone. In response, the first microphone may generate first audio data. Later, when the user would like to be identified by the system, the user may state the utterance again, although this time to a second device having a second microphone. This second microphone may accordingly generate second audio data. Because the acoustic response of the first microphone may differ from the acoustic response of the second microphone, however, this disclosure describes techniques to apply a relative transfer function to one or both of the first or second audio data prior to comparing these data so as to increase the recognition accuracy of the system.