Google Voice Search Attack on Android [pdf]
arxiv.org
arxiv.org
I just don't see this as an effective means of controlling a phone, when your primary means of "infection" involve user action.
The paper even acknowledges this, though almost dismissively: "One potential limitation of GVS-Attack is that the victim may notice the voice command played by VoicEmployer and interrupt it."
That's like, the whole reason this attack isn't very useful except in specific situations. No one's going to be running their botnet via GVS-Attack, that's for sure.
Your idea, however, is exactly the "missing piece" to this attack.
I could see it becoming a serious concern in places like China, where 3rd party markets are much more popular, and republishing existing apps with malware is commonplace.
> Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
Is sharing to HN considered in the same vein as redistributing to a list?
http://www.ianww.com/voice_assistant_abuse.html
Seems like this attack could be trivially prevented by disabling speech recognition while speech synthesis is running. It looks like this may already be the case for the web speech api (or the synthetic voice is just too different from mine).
The paper mentions that they save the synthetic voice to .wav files first, and only later play them back, so this would not help.