ADSactly Tech News: What does AI, Tacotrons and Bitcoin Explainer Videos Have in Common?
AI is developing at a rapid pace and the rate of growth is exponential as advancements and research is incorporated into future experiments and creations.
Not all AI's are intelligent as proven by a recent attempt to use the technology in video production.
It never ceases to amaze me how quickly technology advances. I remember when my parents bought an Apple Macintosh computer when I was in Primary school. It was the coolest thing ever! It had games like Star Wars and bricks! It could even access the internet with a dial-up modem! These were glorious days for technology as kingpins like Steve Jobs and Bill Gates were just getting started transforming our world through innovation! But looking at where we are now, all those cool systems of the past, they seem like toys compared to what is available today!
The term ‘Artificial Intelligence’ didn’t even exist in my lexicon when I was growing up. Today I hear the term splattered all across the internet, in cafes and when I open a newspaper (yes they still exist but just barely!). The speed and pace of technological advancements being made by the world's top scientists and computer engineers in the field of AI will simply shock and amaze you.
In this article, I will explain why artificial intelligence is moving beyond recognition and that fact is demonstrated by recent developments coming out of Google's DeepMind program.
First off, if you've ever used Google products, you may or may not be familiar with Google's AI voice…
There are two settings, one for male and one for female voice and when you hear it you'll know something is majorly off with unnatural pronunciation similar to that of Apple's Siri or Microsoft's Cortana.
Even though this technology has ceased to impress on many occasions, every year the unnatural sound that computer generated voices make is becoming less recognizable.
Just last year Google made a breakthrough and developed a technology sounds much less robotic and much more like a human voice! In fact, the new ‘Tacotron 2’ Google voice AI is almost undistinguishable from a human voice!
You won’t believe this! In fact, I won’t have to work very hard to convince you why it is such a groundbreaking advancement of modern text to speech AI functionality! All you need to do is click play and see for yourself:
In a research paper published by Google, the ‘Tacotron 2’ team introduces details regarding the impressive speech system. In the paper, Google highlights the system’s ability to speak almost identically to its human creators.
The team describes the second-generation speech system in the report stating, "The Tacotron 2 is a neural network architecture for speech synthesis directly from text. The system is composed of a recurrent sequence-to-sequence feature prediction network that maps character embeddings to mel-scale spectrograms, followed by a modified WaveNet model acting as a vocoder to synthesize timedomain waveforms from those spectrograms."
But I can summarize it all in just one word… Wow!
According to the report, the technology that powers this new text to speech tool consists of two deep neural networks. Basically, the way it works is that the first network translates the text into a spectrogram. After the spectrogram has been made it is sent to WaveNet, a new system specially designed by Google's Deep Mind AI program. The process results in a computer generated voice that sounds like its human counterparts.
But digging a bit deeper, I have to ask the question, do we really want AI to have voices that sound identical to those of human beings? How will we know who or what we are talking to when we take the next call from the mystery number? These are the same problems of the famous movie ‘Blade Runner,’ a move which I absolutely love still to this day!
Well now that I’ve asked a philosophical question and took a bit of time to reflect, let’s move on shall we? Google and employees of their DeepMind program have been busy indeed developing AI systems that are already starting to be incorporated into our real, everyday world.
But it is important to note for every accomplishment a few failures needed to be made first!
There are other’s developing AI systems and not everyone paints the Mona Lisa on their first try…
Recently, an AI competition was held with a goal of creating “basic stage training” based on pre-existing Bitcoin (BTC) explainers but sadly, it resulted in almost complete rubbish. The story is quite interesting so I thought readers might get a good laugh out of this one!
See the explainer that was made by AI is incredibly funny because it looks professional but makes absolutely no sense at all! In fact, it doesn't do a good job of explaining Bitcoin at all! Moreover, it actually results in leaving viewers even more confused about Bitcoin then they were before watching the video! All that was needed was to change the competition a bit and say it was about creating an explainer that further confuses a viewer who is trying to learn more about Bitcoin and it would have worked perfectly!
Take a look and see for yourself if I'm right:
Well folks if you made in through the entire video, I’m impressed. In fact after watching it myself, I feel as if my brain has been scrambled and served up with some bacon and baked beans.
I would not recommend having a friend or family member watch this video unless they were already familiar with Bitcoin because it does have the potential to create brand new misconceptions.
Let’s assume you took it at face value and went around networking and trying to explain your understanding of Bitcoin based on this video to new contacts, I’m going to take a shot in the dark and say that you probably wouldn't get very far!
What this does prove however is that in regard to AI, every two steps forward more often than not is proceeded by one step backwards! Yet the technology continues to advance and it frightens the living daylights out of me... I mean as much as anything could frighten a grown man I guess.
Google recently developed another technology even more creepy than the Tacotron in my opinion. If you haven’t seen videos of the new Google Assistant you may be surprised to see how lifelike it can actually be.
I was shocked and amazed to see Google Assistant make an appointment with a girl on the other end of the phone and see her take down the booking while being completely unaware she was talking to AI.
Clearly the Tacotron2 and the new Google Assistant have something in common. It is my belief that each advancement leads us to an even greater advancement and we may be going down a road that is simply unpredictable in nature yet we have little to no ability to slow down.
Is AI dangerous for humanity?
Absolutely.
Are there potential benefits for humanity if we are to advance this technology?
I do believe there are hundreds or even thousands of potential use cases that will prove helpful to human beings and improve our quality of life.
Should we stop and analyze what is going on here so we can take action if deemed necessary to rise up against big corporations who may potentially use this technology to further enslave us? Of course we should.
I know topics like this can be a bit unsettling for folks. I plan to take a break from some of these fringe technology topics in the near future and maybe focus on friendlier technologies as topics for future articles.
But, the fact remains, I would really like to know how my fellow @ADSactly society members feel about these new AI systems coming online, systems like Google Assistant that has the ability to call you and make a booking without you even knowing its a robot.
What did you think of the Bitcoin explainer video? Does it put your mind at ease and give you the impression that maybe we don't have as much to worry about in terms of an AI takeover scenerio as I may have implied at the beginning of this article?
Here's a chance for the @ADSactly community to leave their thoughts and opinions on this topic.
Thanks for reading.
In-text citations sources:
Google’s DeepMind develops AI that can render 3D objects from 2D pictures – VentureBeat
Google's New Voice AI is Hyper Realistic - Interesting Engineering
AI-Based Youtube Bitcoin Explainer Trained By Real BTC Guides Gets It All Wrong - CoinTelegraph
Image Source:
Interesting Engineering
Pixabay
Click on the coin to join our Discord Chat
for Steem witness!Witness proposal is here:
Go To Steem Witness Page
In the bottom of the page type: adsactly-witness and press vote.
Use small letters and no "@" sign. Or, click here to vote directly!
Thank you!