Retail Scale Speech Recognition for Product Info

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Retail scales are typically inactive until the end of the sales process, limiting customer access to product information until the identification number is entered, and fail to provide engaging experiences beyond weight and price display.

Innovation Solution

Integration of a microphone device for speech recognition, allowing real-time conversion of customer-seller conversations into electrical signals, comparing these with stored keywords to output relevant product information via a control device, enabling early information provision and tailored advertising during the sales process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If speech recognition and keyword comparison are integrated into the retail scale, then customer information provision is improved during the sales process, but device complexity increases

Engineering Contradiction:
Improveproduct information availabilityVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The retail scale is enhanced with multiple functional modules including microphone device for speech capture, speech recognition device for converting speech to text, comparison device for keyword matching, and control device for managing information output. This multi-functional integration allows the scale to perform both traditional weighing functions and new information provision functions simultaneously.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

A processor unit serves as an intermediary component that coordinates between the microphone device, speech recognition device, comparison device, and control device. The processor unit manages the flow of information and controls the timing of information output based on recognized keywords, thereby managing system complexity through centralized coordination.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If information is output immediately upon keyword recognition, then customer engagement is improved, but unintentional information output may occur

Engineering Contradiction:
Improveinformation delivery efficiencyVSAvoidinformation output accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system activates a time window before actually outputting information. When a keyword is recognized, the control device starts a predefined time window during which additional keywords must be recognized to trigger information output. This preliminary action allows the system to verify the customer's intent before delivering information, reducing false positives.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from multiple keyword recognitions within a time window to confirm information output. The control device monitors whether additional keywords are recognized during the active time window and adjusts information output accordingly. This feedback mechanism ensures that information is only provided when the customer's intent is clearly indicated.

Inventive Principle:
Principle #23Feedback

3Ease of operation

If speaker-independent speech recognition is used, then ease of operation is improved, but speech recognition accuracy may decrease

Engineering Contradiction:
Improvecustomer interaction convenienceVSAvoidspeech recognition accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The system uses keyword spotting rather than full speech recognition. Instead of attempting to accurately transcribe and understand complete sentences, the system searches for specific predefined keywords in the speech signal. This partial recognition approach maintains ease of operation while reducing the complexity and accuracy requirements of the speech recognition system.

Inventive Principle:
Principle #16Partial or excessive action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enhances customer experience by providing product-related information and advertising at the beginning of the sales process, reducing unintentional information output through time window activation and speaker-independent speech recognition, and tailoring content based on customer demographics.

Implementation Method 1

a microphone device for listening to a conversation between a customer and a salesperson to convert coherent spoken words of the conversation into an electrical voice signal

Methodology Applied
Scientific EffectAcoustic-to-electrical transduction:

Data Source

PatentEP1983493B1Shop scale for processing purchases
Publication Date: 2012.05.16 BIZERBA GMBH & CO KG
  • EP1983493B1 patent drawingFigure 1
  • EP1983493B1 patent drawingFigure 2

AI summary

The commercial weighing machine has a microphone device (11) for listening conversation between a customer and a salesman for converting connected spoken words of conversation into electric speech signal. A control device (13) is arranged to output an information available over a data source (17) or display the information, particularly in the form of selection menu, by a display device (19) by locating a key word and a permissible combination of key words belonging to a group, which has a key word belonging to the group assigned to the located key word or located permissible combination.