There’s more AI news out there than anybody can potentially stay up to date with. You can remain tolerably up to date on the most interesting developments with this column, which collects AI and maker learning developments from around the world and discusses why they may be crucial to tech, startups or civilization.

To start on an easy going note: The methods scientists find to apply maker finding out to the arts are always fascinating– though not constantly practical. A team from the University of Washington wanted to see if a computer system vision system might learn to tell what is being used a piano just from an overhead view of the keys and the player’s hands.

Audeo, the system trained by Eli Shlizerman, Kun Su and Xiulong Liu, sees video of piano playing and first extracts a piano-roll-like basic series of essential presses. Then it includes expression in the form of length and strength of journalisms, and last but not least polishes it up for input into a MIDI synthesizer for output. The results are a little loose however absolutely identifiable.

Diagram showing how video of a piano player's hands on the keys is turned into MIDI sequences.

Image Credits: Shlizerman, et. al “To develop music that sounds like it might be played in a musical performance was previously thought to be difficult,”said Shlizerman. “An algorithm requires to determine the hints, or’ functions,’ in the video frames that relate to creating music, and it requires to’ envision ‘the noise that’s occurring in between the video frames. It requires a system that is both accurate and creative. The reality that we achieved music that sounded pretty good was a surprise.”

Another from the field of letters and arts is this very interesting research into computational unfolding of ancient letters too delicate to deal with. The MIT group was looking at “locked” letters from the 17th century that are so elaborately folded and sealed that to get rid of the letter and flatten it may completely damage them. Their method was to X-ray the letters and set a new, advanced algorithm to work analyzing the resulting imagery.

Diagram showing x-ray views of a letter and how it is analyzed to virtually unfold it.

< img aria-describedby="caption-attachment-2121182"loading="lazy "class="breakout size-full wp-image-2121182"src="https://techcrunch.com/wp-content/uploads/2021/03/unfolding-mit.jpg"alt ="Diagram revealing x-ray views

of a letter and how it is analyzed to virtually unfold it.” width=”1024″height =” 875 “srcset=”https://techcrunch.com/wp-content/uploads/2021/03/unfolding-mit.jpg 1351w, https://techcrunch.com/wp-content/uploads/2021/03/unfolding-mit.jpg?resize=150,128 150w, https://techcrunch.com/wp-content/uploads/2021/03/unfolding-mit.jpg?resize=300,256 300w, https://techcrunch.com/wp-content/uploads/2021/03/unfolding-mit.jpg?resize=768,657 768w, https://techcrunch.com/wp-content/uploads/2021/03/unfolding-mit.jpg?resize=680,581 680w, https://techcrunch.com/wp-content/uploads/2021/03/unfolding-mit.jpg?resize=50,43 50w”sizes=”(max-width: 1024px)100vw, 1024px “> Diagram revealing X-ray views of a letter and how it is analyzed to virtually unfold it. Image Credits: MIT”The algorithm winds up doing an outstanding job at separating the layers of paper, despite their severe thinness and tiny gaps in between them, in some cases less than the resolution of the scan,”MIT’s Erik Demaine said.”We weren’t sure it would be possible.”The work may be applicable to numerous type of documents that are difficult for easy X-ray strategies to unravel. It’s a little a stretch to categorize this as “machine learning,” however it was too intriguing not to consist of. Read the full paper at Nature Communications.

Diagram showing reviews of electric car charge points are analyzed and turned into useful data.

Image Credits: Asensio, et. al You reach a charge point for your electric vehicle and find it to be out of service. You might even leave a bad review online. Thousands of such evaluations exist and constitute a possibly very beneficial map for municipalities looking to expand electrical vehicle infrastructure.

Georgia Tech’s Omar Asensio trained a natural language processing design on such reviews and it soon ended up being a specialist at parsing them by the thousands and ejecting insights like where outages were common, relative expense and other elements.

Article curated by RJ Shara from Source. RJ Shara is a Bay Area Radio Host (Radio Jockey) who talks about the startup ecosystem – entrepreneurs, investments, policies and more on her show The Silicon Dreams. The show streams on Radio Zindagi 1170AM on Mondays from 3.30 PM to 4 PM.