Fooling Speech Recognition With Hidden Voice Commands

It’s 2018, and while true hoverboards still elude humanity, some future predictions have come true. It’s now possible to talk to computers, and most of the time they might even understand you. Speech recognition is usually achieved through the use of neural networks to process audio, in a way that some suggest mimics the operation of the human brain. However, as it turns out, they can be easily fooled.

The attack begins with an audio sample, generally of a simple spoken phrase, though music can also be used. The desired text that the computer should hear instead is then fed …read more http://pje.fyi/QBNwjK

Paul Jacob Evans

Advertisements

The Grafofon: An Optomechanical Sequencer

There are quick hacks, there are weekend projects and then there are years long journeys towards completion.  [Boris Vitazek]’s grafofon falls into the latter category. His creation can best be described as electromechanical sequencer synthesizer with a multiplayer mode. The storage medium and interface for this sequencer is a thirteen-meter loop of paper that is mounted like a conveyor belt. Music is composed by drawing on the paper or placing objects on it. This is usually done by the audience and the fact that the marker isn’t erased make the result collaborative and incremental.  These ‘scores’ are read by a

…read more http://pje.fyi/Px9HGg

Paul Jacob Evans

The Best Stereo Valve Amp In The World

There are few greater follies in the world of electronics than that of an electronic engineering student who has just discovered the world of hi-fi audio. I was once that electronic engineering student and here follows a tale of one of my follies. One that incidentally taught me a lot about my craft, and I am thankful to say at least did not cost me much money.

It must have been some time in the winter of 1991/92, and being immersed in student radio and sound-and-light I was party to an intense hi-fi arms race among the similarly afflicted. Some …read more http://pje.fyi/PgPQb1

Paul Jacob Evans

Bessel Filter Design

Once you fall deep enough into the rabbit hole of any project, specific information starts getting harder and harder to find. At some point, trusting experts becomes necessary, even if that information is hard to find, obtuse, or incomplete. [turingbirds] was having this problem with Bessel filters, namely that all of the information about them was scattered around the web and in textbooks. For anyone else who is having trouble with these particular filters, or simply wants to learn more about them, [turingbirds] has put together a guide with all of the information he has about them.

For those who …read more http://pje.fyi/PbSJF0

Paul Jacob Evans

Designing the Atom Smasher Guitar Pedal

[Alex Lynham] has been creating digital guitar pedals for a while and after releasing the Atom Smasher, a glitchy lo-fi digital delay pedal, he had people start asking him how he designed digital effects pedals rather than analog effects. In fact, he had enough interest, that he wrote an article on it.

The article starts with some background on [Alex], the pedals he’s built and why he chose not to work on pedals full-time. Eventually, the article gets to the how [Alex] designed the Atom Smasher. He starts by describing the chip he used, the same one that many hobbyists, …read more http://pje.fyi/PZCtmW

Paul Jacob Evans

Robotic Glockenspiel and Hacked HDD’s Make Music

[bd594] likes to make strange objects. This time it’s a robotic glockenspiel and hacked HDD‘s. [bd594] is no stranger to Hackaday either, as we have featured many of his past projects before including the useless candle or recreating the song Funky town from Old Junk.

His latest project is quite exciting. He has incorporated his robotic glockenspiel with a hacked hard drive rhythm section to play audio controlled via a PIC 16F84A microcontroller. The song choice is Axel-F. If you had a cell phone around the early 2000’s you were almost guaranteed to have used this song as a ringtone …read more http://pje.fyi/P3HCgp

Paul Jacob Evans

Neural Network Composes Music; Says “I’ll be Bach”

[carykh] took a dive into neural networks, training a computer to replicate Baroque music. The results are as interesting as the process he used. Instead of feeding Shakespeare (for example) to a neural network and marveling at how Shakespeare-y the text output looks, the process converts Bach’s music into a text format and feeds that to the neural network. There is one character for each key on the piano, making for an 88 character alphabet used during the training. The neural net then runs wild and the results are turned back to audio to see (or hear as it were) …read more http://pje.fyi/Ngw9rp

Paul Jacob Evans