Mobile ASR Privacy Protection via Local Obscuring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic speech recognition (ASR) systems on mobile devices face challenges in protecting private information during client-server processing, as they often transmit sensitive data to remote servers without adequate protection, potentially compromising user privacy.
Innovation Solution
A mobile device with a local ASR arrangement identifies and obscures private information by creating hash codes or removing/replacing it before transmitting speech input signals to a remote server, ensuring that private data is not shared, and reinserts the obscured data into recognition results for user display, thereby maintaining privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If private information is transmitted to remote server for ASR processing, then speech recognition accuracy is improved, but user privacy is compromised
Solution Approach 1:
The patent extracts and identifies private information from the speech input signal before transmission to the remote server. The local ASR arrangement determines if private information is included and obscures it, effectively removing the harmful element while allowing the remaining signal to be processed remotely for maintained accuracy.
Solution Approach 2:
The patent introduces an intermediary obscuring mechanism between the speech input and the remote server. This intermediary component (the control unit that obscures private information) allows the signal to be transmitted while protecting privacy, acting as a mediator that enables both remote processing and privacy protection.
2Object-affected harmful factors
If local ASR processing is performed to identify private information, then privacy protection is improved, but device complexity increases
Solution Approach 1:
The local ASR arrangement on the mobile device performs multiple functions: it processes speech input, identifies private information, and obscures it before transmission. By making the local ASR component multi-functional, the patent avoids adding separate dedicated privacy protection hardware, thereby limiting the increase in device complexity while maintaining strong privacy protection.
3Object-affected harmful factors
If private information is obscured using hash codes, then privacy protection is improved, but data accuracy for recognition is reduced
Solution Approach 1:
The patent segments the speech input signal into private information portions and non-private portions. Only the private portions are obscured with hash codes while the non-private portions remain intact and are transmitted to the remote server. This segmentation allows the system to maintain data accuracy for recognition while protecting privacy for sensitive segments.
Data Source
AI summary
A mobile device is adapted for protecting private information on the mobile device in a hybrid automatic speech recognition arrangement. The mobile device includes a speech input component for receiving a speech input signal from a user. Additionally, the mobile device includes a local ASR arrangement for performing local ASR processing of the speech input signal and determining if private information is included within the speech input signal. A control unit on the mobile device obscures private information in the speech input signal if the local ASR arrangement identifies information within a speech recognition result as private information. The control unit releases the speech input signal with the obscured private information for transmission to a remote server for further ASR processing.Results from the remote server's ASR processing are integrated and combined with results from local ASR processing to display information on the mobile device.