A developer named Diego Trujillo Pisanty from Mexico has created a Raspberry Pi-powered device that can make images out of the sounds you feed in.
This blind camera project is so limited but unimaginable with this hardware. Pisanty used Raspberry Pi 3B, Python 3 and his own artificial neural network – all bound in a plastic shell with a horn to receive sound.
An Audio Camera From Pi and Python
The rise of AI led to the growth of several intriguing inventions recently, with ChatGPT being the top scorer. Well, a maker and developer from Mexico named Diego Trujillo Pisanty has invented the new camera that can make images from sound inputs!
This was showcased in his YouTube channel, where he held a box with a horn attached to receive sound inputs. He coded the program to convert these sounds into images in Python 3, based on a Raspberry Pi chip. Pisanty developed his own artificial neural network (or ANN) for this project, which is trained on a set of videos taken around Mexico City.
And since it was based on the inner-city images of Mexico City, be prepared to see the output related to Mexican tastes. The built-in AI will parse the sound inputs to create images, says Pisanty.
Though Pisanty used Raspberry Pi 3B to make this project, he says that Raspberry Pi 4 too can be used. All in all, it should be able to handle Tensorflow, where an attached screen guides the handler. It acts as a viewfinder and shows the preview of images created. For more details, you may check Pisanty’s YouTube channel and his website.