Security vulnerabilities in speech recognition systems
The aim of this project is to develop a generic methodology for blackbox testing of speech recognition systems by seeking to identify homophones or similar sounding words which may be misrecognised by a system as command words. A series of experiments are performed with two different speech recognit...
| Main Author: | |
|---|---|
| Format: | Report |
| Published: |
Centre for Doctoral Training in Cyber Security
2016
|
| _version_ | 1797086013957341184 |
|---|---|
| author | Bispham, M |
| author_facet | Bispham, M |
| author_sort | Bispham, M |
| collection | OXFORD |
| description | The aim of this project is to develop a generic methodology for blackbox testing of speech recognition systems by seeking to identify homophones or similar sounding words which may be misrecognised by a system as command words. A series of experiments are performed with two different speech recognition systems. The data used in the experiments are a set of words from the vocabulary of a controlled natural language, Attempto Controlled English, together with a set of words not permitted in that controlled language. Several instances are identified where a ‘legal’ Attempto word is misrecognised as an ‘illegal’ word. This demonstrates the feasibility of an attack whereby an attacker might seek to find apparently innocuous words which are misrecognised by a speech-controlled system as commands, thus enabling the attacker covertly to prompt the system to perform an unauthorised action. In such a situation, the attacker would identify a set of command or ‘target’ words used to control a system (the equivalent of the ‘illegal’ words in the experiments with Attempto) and would seek to find a set of ‘adversarial’ words which are misrecognised by the system as a target word. In a real-life attack, an attacker might seek to find words which are misrecognised as command words for a digital assistant such as Siri or Cortana, or as command words for a voice-controlled device in the Internet of Things. |
| first_indexed | 2024-03-07T02:15:59Z |
| format | Report |
| id | oxford-uuid:a241398d-1a65-411c-b385-8526e97caff2 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T02:15:59Z |
| publishDate | 2016 |
| publisher | Centre for Doctoral Training in Cyber Security |
| record_format | dspace |
| spelling | oxford-uuid:a241398d-1a65-411c-b385-8526e97caff22022-03-27T02:18:54ZSecurity vulnerabilities in speech recognition systemsReporthttp://purl.org/coar/resource_type/c_93fcuuid:a241398d-1a65-411c-b385-8526e97caff2Symplectic Elements at OxfordCentre for Doctoral Training in Cyber Security2016Bispham, MThe aim of this project is to develop a generic methodology for blackbox testing of speech recognition systems by seeking to identify homophones or similar sounding words which may be misrecognised by a system as command words. A series of experiments are performed with two different speech recognition systems. The data used in the experiments are a set of words from the vocabulary of a controlled natural language, Attempto Controlled English, together with a set of words not permitted in that controlled language. Several instances are identified where a ‘legal’ Attempto word is misrecognised as an ‘illegal’ word. This demonstrates the feasibility of an attack whereby an attacker might seek to find apparently innocuous words which are misrecognised by a speech-controlled system as commands, thus enabling the attacker covertly to prompt the system to perform an unauthorised action. In such a situation, the attacker would identify a set of command or ‘target’ words used to control a system (the equivalent of the ‘illegal’ words in the experiments with Attempto) and would seek to find a set of ‘adversarial’ words which are misrecognised by the system as a target word. In a real-life attack, an attacker might seek to find words which are misrecognised as command words for a digital assistant such as Siri or Cortana, or as command words for a voice-controlled device in the Internet of Things. |
| spellingShingle | Bispham, M Security vulnerabilities in speech recognition systems |
| title | Security vulnerabilities in speech recognition systems |
| title_full | Security vulnerabilities in speech recognition systems |
| title_fullStr | Security vulnerabilities in speech recognition systems |
| title_full_unstemmed | Security vulnerabilities in speech recognition systems |
| title_short | Security vulnerabilities in speech recognition systems |
| title_sort | security vulnerabilities in speech recognition systems |
| work_keys_str_mv | AT bisphamm securityvulnerabilitiesinspeechrecognitionsystems |