ATM Voice Control via Query Language Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated transaction machines (ATMs) lack the ability to be controlled without the use of a keypad, limiting user interaction and accessibility.
Innovation Solution
A menu crawling system utilizing query language processing (QLP) and natural language processing (NLP) circuitry allows for voice-operable control of ATMs by translating voice or text commands into control signals to navigate the tree-structured menu, enabling users to perform functions like withdrawals and deposits without physical input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If voice-operable control is implemented without keypad, then user accessibility and interaction ease are improved, but device complexity increases due to addition of NLP and QLP circuitry
Solution Approach 1:
The patent introduces NLP circuitry and QLP circuitry as intermediary components between the user's voice commands and the ATM's existing menu system. The NLP circuitry processes natural language input, while the QLP circuitry translates queries into menu navigation commands, effectively mediating between the new voice interface and the legacy ATM system without requiring complete system redesign
Solution Approach 2:
The patent replaces the mechanical keypad input system with a voice-based acoustic input system. Users speak commands instead of physically pressing buttons, substituting mechanical interaction with acoustic field interaction processed through microphones and speech recognition algorithms
2Productivity
If menu crawling system is implemented, then transaction processing speed is improved, but device complexity increases due to additional processing circuitry
Solution Approach 1:
The system performs preliminary actions by pre-processing voice commands through NLP circuitry before executing menu navigation. The QLP circuitry pre-translates user intent into structured query commands that can be efficiently executed by the ATM's existing menu system, reducing real-time processing requirements during actual transactions
3Ease of operation
If natural language processing is used instead of keypad input, then user accessibility for disabled individuals is improved, but measurement precision of user intent may deteriorate
Solution Approach 1:
The patent implements feedback mechanisms where the system processes natural language queries through NLP and QLP circuitry, then provides responses that allow users to confirm or correct interpretations. This feedback loop enables refinement of command accuracy while maintaining the accessibility benefits of voice input
Solution Approach 2:
The system changes the parameter of input modality from discrete keypad presses to continuous natural language speech. By transforming rigid button selections into flexible speech-based queries processed through NLP, the system accommodates diverse user abilities while using algorithmic processing to maintain precision in intent recognition
Data Source
AI summary
Systems, apparatuses, methods, and computer program products are disclosed for controlling a function of an automated transaction machine (ATM). An example method includes receiving, by a query language processing (QLP) circuitry, a query command signal. The example method further includes generating, based on the query command signal and a tree-structured menu of an ATM, a set of control signals configured to instruct the ATM to perform the function by crawling a tree-structured menu of the ATM. Subsequently, the example method includes transmitting, to an ATM, the set of control signals to cause display of the advancement of the tree-structured menu. Additionally, the example method includes a text to speech (TTS) circuitry configured to receive a response signal and generate, based on the response signal, electronic information indicating a vocal representation of the response signal. The example method may further include transmitting to the ATM, the electronic information.