Securing Speech Recognition Data via Encryption Key Management

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Client/server speech recognition systems face security vulnerabilities due to the risk of unauthorized access to sensitive information, such as personal health data, when it is transmitted and stored in unencrypted form, making it susceptible to malicious attacks.

Innovation Solution

Implementing encryption and decryption key management to limit access to sensitive information, where encryption keys are generated and managed by client computers and servers, ensuring that decryption keys are not transferred unencrypted, and encrypting speech recognition input and output data to protect it during transmission and storage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If speech recognition systems process unencrypted sensitive information, then processing efficiency and accessibility are improved, but security and vulnerability to malicious attacks deteriorate

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidvulnerability to malicious attacks

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The system performs encryption before data transmission and decryption before processing, preparing data in advance to prevent security vulnerabilities during transmission and storage while maintaining processing efficiency when data is in decrypted form

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Encryption and decryption keys act as intermediaries that enable secure data transformation, allowing the system to maintain data security during transmission while enabling efficient processing when data is temporarily decrypted in controlled environments

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If decryption keys are transferred to enable speech recognition processing, then data accessibility and processing capability are improved, but security risk from key interception deteriorates

Engineering Contradiction:

Improvedata accessibilityVSAvoidrisk of key interception

Core Design Contradiction:

Ease of operationVSObject-affected harmful factors

Solution Approach 1:

Decryption keys are generated and distributed in advance through secure channels before data transmission, enabling the system to establish secure access capabilities beforehand while minimizing the window of vulnerability during key transfer

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses temporary decryption keys that are generated, used, and discarded for each processing session, preventing long-term key storage and reducing the risk of key interception and reuse by malicious actors

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

3Reliability

If encrypted data is stored and transmitted, then security against unauthorized access is improved, but processing complexity and computational overhead increase

Engineering Contradiction:

Improvesecurity against unauthorized accessVSAvoidprocessing complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments data handling into distinct encrypted storage/transmission phases and decrypted processing phases, allowing security and processing complexity to be separated in time and space, with encryption/decryption operations performed only when necessary

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system replaces complex continuous encryption/decryption operations with discrete, event-driven cryptographic transformations that occur only at specific transition points (storage, transmission, processing), reducing overall computational overhead while maintaining security

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS9032219B2Securing speech recognition data

Publication Date: 2015.05.12 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Methods and apparatus for reducing security vulnerabilities in a client/server speech recognition system including one or more client computers and one or more server computers connected via a network. Decryption of sensitive information, such as medical dictation information, is performed on designated servers to limit the attack surface of unencrypted data. Management of encryption and decryption keys to restrict the storage and/or use of decryption keys on the server side of the client/server speech recognition system, while maintaining encrypted data on the server side is also described.