How an empty crisp packet can be used to eavesdrop on conversations

http://www.theguardian.com/technology/2014/aug/05/mit-crisp-packet-eavesdrop-audio-sound

According to an article on technology published in The Guardian by Alex Hern, researchers at the Massachusetts Institute of Technology (MIT) are conjoining with major companies like Microsoft and Adobe in order to construct a "method of reconstructing sound from a video of an object". With this, instead of using expensive equipment, the researchers just have to observe with a very high speed camera the vibrations that sound gives off on everyday objects, like and empty crisp bag. Sound emits waves that cause vibrations in objects, and although invisible to the naked eye, it can be easily captured. The premise of this article is that if you have the equipment required -- which is not too hard to have, given that a regular rolling shutter camera can practically do the job -- then you are able to listen to every sound being emitted even if it is recording mutely. This already poses a question on morality, on whether this is another way that the government and people "above the law" will use to spy on regular people. Although these results must be taken with a grain of salt, if the experiment is stripped down, it basically means that whoever has certain equipment will be able to overhear conversations and recognize about every type of sound wave being emitted. The article states that "As well as the obvious applications for surveillance and law enforcement, [the researchers plan] on seeing whether the acoustic properties of objects reveal more information about their internal make up, calling it “a new kind of imaging”". To me this sounds very skeptical, because a lot of important information can be revealed through this kind of technology and can be put to bad use a lot easier than it could be put to helping understand more about the scientific part of things. People who use this technology could even be considered to be in a different "world" through a ToK perspective because they will be able to acquire information in a way that regular people cannot, listening and understanding through the use of images. This article reminded me of sense perception because it is basically talking about the privilege some people may have now that will allow one sense to replace the other, allowing visuals to become sounds. At least from my level of understanding of the topic, there are limitations that the article does not address, one of them being that the interference of background noise and ambience can very much skew the accuracy of the collected information. This already partially invalidates the article in a sense because the article has a strong position on only informing how well this will work and all the ways it can be executed, but it ceases to speak of the accuracy. This technology could be potentially dangerous, because extracting sounds merely based on visual vibrations on an image does not seem very reliable and is very much based on deduction, and other technologies can do a much better job on capturing more clean audio. People may assume a lot of things that are extracted from the audio that may not be exactly what actually happened, and that can lead to many issues, especially if this technology is used for governmental experiments.