Voice Privacy: Analyzing Trends and Patterns inAmazon Alexa Voice History; The Struggle for Control of Personal Data

Author:
Hancock, Matthew, School of Engineering and Applied Science, University of Virginia
Advisors:
Tian, Yuan, EN-Comp Science Dept, University of Virginia
Norton, Peter, EN-Engineering and Society, University of Virginia
Abstract:

Internet connected voice assistants like Alexa, Google Assistant, and Siri sometimes record unwanted conversations. Deleting them is complicated and preventing them is even more difficult.

How can trends and patterns in Amazon Alexa voice history aid owners in tracking recorded conversations and deleting unwanted voice data? The research team developed a Google Chrome extension to download transcripts and audio recordings from Alexa voice history, while another research team developed a system to quickly identify unintended recordings. Future researchers will combine projects from both teams to analyze voice data and find methods to minimize unintentional Alexa recordings.

In the U.S. since 2010, how have privacy advocates attempted to limit data collection by smart home devices? Privacy advocates believe tech companies have too much control over personal data. They fear how it is used by corporations, the government, and hackers. As the market for smart home devices grows, tech companies face rising pressure from privacy advocates to secure their products and better manage user data. Because privacy can impair functionality, tech companies are reluctant to adapt.

Degree:
BS (Bachelor of Science)
Keywords:
Voice History, Voice Privacy, Amazon Alexa, Amazon Alexa Chrome Extension
Notes:

School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Yuan Tian
STS Advisor: Peter Norton
Technical Team Members: Gabriel Simmons, Tu Le, Danny Huang

Language:
English
Issued Date:
2021/05/08