News for hackers. Stuff that matters.

A comment on Hackers send silent commands to speech recognition systems with ultrasound. Parent comment

Re:Hackers send silent commands to speech recognition system... by rmonroe on Friday September 08, 2017 @08:21AM (#15199634)
They are taking advantage of nonlinear properties in the microphone system, as described clearly in the article. There is no software filtering to protect against this using classical signal processing.

That said, it may be possible to detect the difference between human speech and the ultrasonic trick with a machine learning solution.

[ Parent | on HN ]
Re:Hackers send silent commands to speech recognition system... by delinka on Friday September 08, 2017 @09:43AM (#15200064)
The thing is, this is creating harmonic vibrations in the membrane at human speech frequencies. I really don't think you need to go with ML. It seems to me that if you remove the LPF, let the ADC access all the frequencies, then (either in the ADC or in software) detect the harmonics and ignore those with an ultrasound source.

Edit: suggest LPF removal.

[ Parent | on HN ]
Re:Hackers send silent commands to speech recognition system... by vel0city on Friday September 08, 2017 @10:00AM (#15200180)
Wouldn't it go through a bandpass filter to remove the ultrasonics before it makes it to the ADC? I thought that was common with DACs to reduce aliasing issues.
[ Parent | on HN ]
Re:Hackers send silent commands to speech recognition system... by delinka on Friday September 08, 2017 @10:18AM (#15200305)
Apologies - I was amidst edits when I was called away. I've addressed this by suggesting removal of the LPF.
[ Parent | on HN ]