Facebook looks for more opportunities to bombard you with ads 👓, an AI that tells you what supplements are OK to take 💊, an audio ad blocker and more — This Week in Machine Learning
The first audio ad blocker has been released, and to be honest… it was just a matter of time. I didn’t give it a try yet, but I expect it to be a bit of a resource hog. You can play with it and try it out yourself on whatever sound clips you’d like as the engine behind Adblock Radio is entirely open source!
The tech this is built on tickles our nerdy neurotransmitters quite a lot too. It uses a combination of acoustic fingerprinting (like a constantly-running Shazam) and an LSTM (long short-term memory network) classifier trained on Mel spectrograms. If those words make any sense to you (maybe even get you excited), you should go ahead and read more:
After giving away your personal information and trying to infiltrate your house through an ACTUAL PORTAL, Facebook seems set on building a new layer on top of reality through their AR Glasses and LiveMaps.
I guess we’ll soon need an AR ad blocker too…
Oh, and to accelerate your merger into the matrix even more, Facebook has just acquired Ctrl-labs a startup working on brain-machine interfaces. What a time to be alive… and not fully digital yet…
All of this tech does get me incredibly excited. But then I remember that’s Facebook behind all of it that excitement slowly transforms into angst. Let’s hope they can redeem themselves and win our trust once more as the tech is coming whether we like it or not!
Talking about glasses, Amazon launched their Alexa-powered smart glasses. I still feel awkward talking to digital assistants in public… but yeah… This is a thing now.
Remember how IBM’s Watson was set to revolutionize health by having its AI read all of the state of the art papers, synthesize their knowledge and always be up to date with the latest and greatest? Well, that monumentally under-delivered, but a similar tool seems to emerge recently, built on top of the years of research behind Semantic Scholar by Allen Institute for AI.
Its name is supp.ai. Its only goal is to stay up to date with literature regarding dietary supplements and how they interact with various conditions and other substances democratizing access to medical information for the average user. Give it a try here:
Startup Spotlight
You probably don’t have a use for the app we have in our Startup Spotlight this week… We don’t even know if it’s a startup or not. Yet we love the idea of machine learning being applied in agriculture! Anything that increases yield while lowering the amount of pesticides is brilliant!
This is exactly what Tumaini (hope in Swahili) does for bananas (told you). If you’re interested to know how they did it and how you can do it yourself for your own plants, head over to this blog post — or wait until Cognifeed releases ;)
If you have a banana tree lying around, you can also give it a try by downloading their Android app from the Play Store:
Before we go, we’d like to remind you that we released a new article on how you can use Cognifeed in influencer marketing to target those influencers that align the best with your brand’s values. The first part is about scraping Instagram for the data you need, check it out below!
That’s it for this week! Until next one, have a good one!
