Remote End Input Validation for IVR Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional IVR systems are inefficient due to high dependency on Speech Recognition Servers, leading to increased call duration, higher resource and bandwidth usage, and complex administration, as they require multiple iterations with the server for user input validation and often necessitate additional licenses to handle peak loads.
Innovation Solution
Implementing remote end input validation using distributed speech recognition capabilities on user devices, where a Request Voice Data Capsule is sent to the device for input validation, reducing the need for continuous communication with the IVR system and utilizing local resources for validation, thereby enhancing efficiency and reducing bandwidth usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the IVR system uses centralized Speech Recognition Server for input validation, then validation accuracy is maintained, but call duration increases and system availability risk increases
Solution Approach 1:
The patent divides the speech recognition function into two segments: centralized grammar/validation rules stored on the IVR server, and local speech-to-text processing on the user device. This segmentation allows validation to occur locally without continuous server communication, reducing call duration while maintaining accuracy through centralized rule enforcement.
Solution Approach 2:
The system performs preliminary action by downloading grammar files and validation rules to the user device before the call occurs. This pre-loading of validation criteria enables the device to perform local speech recognition and validation without real-time server dependency, thus reducing call duration while preserving validation accuracy.
2Stability of the object's composition
If the IVR system relies on centralized Speech Recognition Server, then validation consistency is ensured, but system availability risk and complexity increase
Solution Approach 1:
By segmenting the speech recognition system into local processing (on user device) and centralized rule storage (on IVR server), the patent reduces single-point-of-failure risks. The local device can perform validation independently using pre-downloaded grammars, ensuring system availability even when server connectivity is interrupted, while consistency is maintained through centralized rule management.
Solution Approach 2:
The user device performs self-service by executing local speech recognition and validation using embedded resources and pre-downloaded grammar files. This self-sufficient operation reduces dependency on the centralized server for each validation operation, improving system availability while maintaining consistency through periodic grammar updates from the server.
3Reliability
If additional Speech Recognition Server licenses are purchased to handle peak loads, then service availability during peak times is improved, but system complexity and cost increase
Solution Approach 1:
The user device performs speech recognition and validation locally using its own processing resources and pre-downloaded grammar files. This eliminates the need for additional server licenses to handle peak loads, as the validation capability resides on the client device. The system scales automatically without requiring proportional increases in server capacity, reducing both complexity and cost.
Solution Approach 2:
By pre-loading grammar files and validation rules to the user device before peak usage periods, the system prepares local validation capabilities in advance. This preliminary action enables the device to handle speech validation independently during peak times without requiring additional server resources, thus improving service availability during high-load periods without increasing system complexity.
4Measurement precision
If multiple iterations with Speech Recognition Server are performed for input validation, then input accuracy is ensured, but bandwidth usage and call duration increase
Solution Approach 1:
The patent segments the validation process into local speech-to-text conversion on the user device and centralized grammar-based validation on the IVR server. This segmentation eliminates the need for multiple iterative communications with the server for each validation step, as the device performs local validation using pre-downloaded grammars, thus reducing bandwidth usage while maintaining input accuracy.
Solution Approach 2:
The system performs preliminary action by downloading comprehensive grammar files and validation rules to the user device before the call. This pre-loading enables the device to perform multiple validation iterations locally without additional server communications, reducing bandwidth usage while ensuring input accuracy through repeated local validation against the pre-loaded grammar rules.
Data Source
AI summary
A method, apparatus and computer program product for providing remote end input validation is presented. A communication is sent to an IVR from a remote end device. The IVR responds by sending a Request Voice Data Capsule (request VDC) to the remote end device. The remote end device receives the Request Voice Data Capsule (request VDC) from the IVR and executes a contained in the remote VDC. The remote end communications device validates user inputs utilizing a resource associated with said remote end communications device and sends a Response Voice Data Capsule (response VDC) including at least one validated communication to the IVR. The IVR processes the response VDC.