English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

This paper investigates the privacy leakage of smart speakers under anencrypted traffic analysis attack, referred to as voice command fingerprinting.In this attack, an adversary can eavesdrop both outgoing and incoming encryptedvoice traffic of a smart speaker, and infers which voice command a user saysover encrypted traffic. We first built an automatic voice traffic collectiontool and collected two large-scale datasets on two smart speakers, Amazon Echoand Google Home. Then, we implemented proof-of-concept attacks by leveragingdeep learning. Our experimental results over the two datasets indicatedisturbing privacy concerns. Specifically, compared to 1% accuracy with randomguess, our attacks can correctly infer voice commands over encrypted trafficwith 92.89\% accuracy on Amazon Echo. Despite variances that human voices maycause on outgoing traffic, our proof-of-concept attacks remain effective evenonly leveraging incoming traffic (i.e., the traffic from the server). This isbecause the AI-based voice services running on the server side responsecommands in the same voice and with a deterministic or predictable manner intext, which leaves distinguishable pattern over encrypted traffic. We alsobuilt a proof-of-concept defense to obfuscate encrypted traffic. Our resultsshow that the defense can effectively mitigate attack accuracy on Amazon Echoto 32.18%.

Fingerprinting Encrypted Voice Traffic on Smart Speakers with Deep Learning