2008
DOI: 10.1109/sp.2008.21
|View full text |Cite
|
Sign up to set email alerts
|

Spot Me if You Can: Uncovering Spoken Phrases in Encrypted VoIP Conversations

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

1
128
0
1

Year Published

2008
2008
2019
2019

Publication Types

Select...
3
3
2

Relationship

0
8

Authors

Journals

citations
Cited by 165 publications
(131 citation statements)
references
References 22 publications
1
128
0
1
Order By: Relevance
“…Recent research has demonstrated that traffic analysis can be used for a number of attacks: identification of web sites visited [3], [4], guesses at passwords typed [5], and recovery of phrases spoken over VoIP [6]. However, it was previously assumed that such attacks required access to one of the routers along the path of Alice's traffic.…”
Section: Introductionmentioning
confidence: 99%
“…Recent research has demonstrated that traffic analysis can be used for a number of attacks: identification of web sites visited [3], [4], guesses at passwords typed [5], and recovery of phrases spoken over VoIP [6]. However, it was previously assumed that such attacks required access to one of the routers along the path of Alice's traffic.…”
Section: Introductionmentioning
confidence: 99%
“…Wright et al [5] propose to use the size of transmitted packets to determine whether a given phrase is spoken. By reducing speech to phonemes, the most basic speech unit, they are able, using HMMs, to reconstruct sentences using the sequence of packet lengths.…”
Section: A Related Workmentioning
confidence: 99%
“…Experiments have shown that the approach proposed in [5] does not work with Skype traffic as the encoding of phoneme depends strongly on the surrounding sounds. Furthermore, the approaches proposed in [5] and [6] require the possession of a binary of the audio codec to create a match between the VBR output and the encrypted packet length.…”
Section: B Motivationmentioning
confidence: 99%
See 1 more Smart Citation
“…In particular, a knowledgeable and powerful censor could potentially defeat such measures by applying steganographic detection techniques [9], enumerating the location of decoy routers [10], and/or leveraging machine learning-based traffic analyzers to perform traffic classification (cf. [14][15][16]). …”
Section: Introductionmentioning
confidence: 99%