A bit of history
In Stanley Kubrik’s movie 2001: A Space Odyssey from 1968 we find the first collision between a computer with a voice and a human being. Of course, the scale of that clash is much more dramatic than the daily struggle to make our voice assistant in the living room understand us, and the sometimes ridiculous misunderstandings that arise when using these extraordinary devices!
In April 2010, Apple acquired a company called Siri which a few months earlier had released an app of the same name for iOS equipped with a speech recognition engine based on machine learning algorithms. With the release of iPhone 4S (October 4, 2011) Siri was integrated into the iOS operating system and has since become an integral part of all Apple products released in subsequent years.
Its reception was lukewarm mainly due to the lack of flexibility and the inability to understand dialect accents. However, everyone appreciated the ability to give voice commands that could interact with the user’s phone, agenda, and e-mails.
Microsoft’s voice assistant Cortana, installed on a Windows Phone 8.1, was first shown at the 2014 Build conference. In November 2014, Amazon announced the arrival of Alexa, which was installed on the first Echo products (here you will find a rundown of all the devices released in the following years). The significant change here is that the voice assistant was just on our phones and our computers anymore, but entered our homes.
Alexa was developed from a product called Ivona, invented in Poland and was also inspired, coincidentally, by Kubrick’s film. The name was chosen because of the presence of the consonant x, easier to recognize for activation (here, our Salvatore Merone explained the difficulties of vocal activation)
The first product developed by Google was announced in May 2016 (here is the video shown at Google/IO) and put on the market in the United States in November 2016 and around the world in 2017. It has since become a product that everyone knows about: Google Home.
Little by little, our homes have filled up with virtual assistants across all brands, prices, and capabilities. The same assistants were made available simultaneously on our phones and tablets. Devices tend to have a low price, sometimes they are given to us as a gift or included for free when buying our televisions. For those who work in the world of technological solutions, it is natural to have these “wonders” (in my case I lost count) as well as all of those devices capable of interacting with them.
We have now entered the era of so-called “Ambient Computing”. But what exactly is it?
The definition is that of an environment (home, workstations, public spaces, hospitals) in which intelligent devices manage data and make decisions based on AI algorithms. These devices are sensitive and responsive to the presence of humans as well as capable of tacitly performing background operations.
What are the pieces to this puzzle?
- Internet of Things (IoT): an ecosystem of devices equipped with sensors capable of collecting environmental data and transmitting them to centralized hubs, perhaps on the cloud, in order to process them all.
- Machine Learning, which we talked about here.
- Conditional Rules: hey allow the devices, thanks to the two previous points, to act in a certain way given the right conditions. It can be a simple rule such as, for example, lifting the shutters in the morning or alerting someone in the house to the presence of a person at the entrance.
- Simple integration: everything must be made available without complex installations and configurations (think for example of Alexa who is able to answer the doorbell even when you are not there).
Let’s see what the latest news and trends are at the beginning of 2022 by analyzing the main producers.
At the re:Invent 2021 conference, there were several presentations dedicated to Alexa. First of all, we have been given some very interesting and exact data that tell us how pervasive Amazon devices really are in our society. There are hundreds of millions of devices around the world that use Alexa.
Ok, we need other metrics because mere possession is not indicative of real user engagement. Applications called skills are installed on Alexa. Are they downloaded? Used? Amazon tells us that the number of interactions with the user is around 50 billion per month and this number is growing at a rate of 40% per year.
Cosa fanno queste skill? Se le dividiamo in categorie troviamo dei risultati What do these skills do? If we divide them into categories, we find surprising results. First of all, there are the musical ones. An independent musician, not necessarily a star, can literally build a direct connection with his fans. Then there are the audio ones, whereby audio we don’t necessarily mean music: sound effects, sounds for sleep, audio mixers (“Alexa, play the meme XXX”). The third fastest-growing category is that of gaming: using your voice to navigate a fantasy world or simply answering quizzes about your favorite team.
There are about 130000 published skills (not counting those developed privately as a hobby) and Alexa developers are also able to monetize their work.
Amazon believes that assistants should not only be found in our living rooms but that they can be installed anywhere they can add value to a business: this is the Alexa Smart properties project. Examples cited are hotels, hospitals, retirement homes, and even individual workstations.
Let’s consider a hotel. A cleaner, on his morning tour, may find that the shower in one room has a plumbing problem. How complex and efficient is the process of reporting the problem to the team of specialists who have to solve it? Isn’t it easier to have a skill that allows you to immediately open a ticket via a voice signal? For more details, I refer you to the following video.
There are around 140,000 smart home products that can be controlled by Alexa (and don’t just think about lamps and thermostats!). Here you can take a look at the ones that have been the most successful.
What are the new devices created by Amazon? Echo Show 10 (4th generation) has a rotating base capable of following the movements of the person who is speaking. Echo Show 15 was also announced, interesting because it could create a new trend: it has, in fact, a 15.6 ”full HD screen, the largest created so far. The last product I want to mention is Astro, an autonomous robot that is an Alexa device capable of moving around the house and communicating with people.
We know that 63% of the total US population uses a voice assistant on a daily basis (NPR AND EDISON RESEARCH: THE SMART AUDIO REPORT). And globally? There are 1.83 billion people using voice assistants (Source: Statista) for a total expenditure of $ 7.3 billion in 2020 (source Juniper).
What direction is Google taking? A clearly ambitious plan was announced in the keynote of the Smart Home Summit 2021 event.
L’attenzione è quindi spostata sugli sviluppatori e sulle aziende che producono The focus has therefore shifted to developers and companies that produce devices for an increasingly intelligent home. In just five years, Google has arrived at products such as Nest, the Google Home app, the use of AI engines on Android, offering an ecosystem to tens of thousands of devices such as washing machines, refrigerators, ovens, light bulbs, air conditioners and even yogurt makers!
The opening of the Google Home Developer Center portal is now near (Spring 2022) where you will find robust and tested software libraries able to be used for your projects on any device. The portal does, however, favor the libraries which will implement the Matter protocol (see next paragraph).
Finally, a real development environment (Google Home IDE) will be available in the form of a Visual Studio Code plugin that will simplify, among other things, the connection to the Google cloud to manage the saving of logs and their analyses.
Google has opened a new protocol for smart home devices with the intention of attracting some of the most important companies.
This originated with the belief that smart home devices should be safe, reliable, and simple to set up. The Egg of Columbus is obviously having an IP protocol that allows communication between devices, mobile apps, and cloud services.
So far, when purchasing these devices, our choice has gone to those that are compatible with the ecosystem created in our home. In the near future, however, a device that follows the Matter specifications will be configurable and manageable through any control system (for example the Google Home or Siri, or Alexa app)
Compared to the competition, Siri will only be made available to third-party devices in 2022. The voice assistant will be redirected through a HomePod device. A demo based on an Ecobee thermostat was shown at the WWDC conference. Full support for Matter has also been announced starting with iOS version 15.
In the absence of its own home assistant (and for some years now also a mobile operating system), Microsoft has decided to relocate Cortana to the market and to close the Cortana app for iOS and Android on March 31st, at the same time interrupting support for all devices that used it, including Microsoft headphones.
Their intention is to integrate the assistant into the Teams mobile app, a product that has exploded in popularity in recent years. At the moment, however, support is only available in the English language. Cortana is still available on the Windows 11 operating system also through the Microsoft To Do app.
The future of ambient computing
With the spread of protocols such as Matter, a strong increase in smart devices is expected in the coming years. How will they change? They will probably be quieter, that is, able to anticipate our requests. We will need fewer voice commands such as “Hey Siri” or “Alexa” and above all, we expect them to be faster, more efficient, and less rigid, since human intervention will not be required to authorize each action.
Jeff Blankenburg, of the Amazon Alexa team, spoke about the possible future scenarios of Ambient Computing in a public event. What do all human beings have in common? The fact that each of us has so many things to do (probably too many) and our brain is constantly busy thinking about it. How can I manage to do all these things? When was your last visit to the ophthalmologist? Did I leave the oven on? If we think of an AI-based personal assistant we would like it to remind us of everything, to anticipate our needs, to offer us all the information we need at exactly the right time, and to allow us to focus on what is really important.
But to get to this result you need to have total trust in it. We have to remove any and all doubts about its reliability, otherwise, our brain will never really get rid of that chaos I was talking about before. The example shown in the video is enlightening: how much trust do you have in your Bluetooth headphones and how much confidence is in the fact that the sun will rise tomorrow? If the headphones don’t work, we aren’t devastated: at the worst, our frustration increases.
Trust is fundamental in the interaction between a human being and a technological device and it will be even more so in the next 5-10 years when we will see if our living room is full of cute gadgets or instead, full of assistants that are really able to help us out with our daily lives.
If you are curious to know how we will use the great potential of the announced tools, keep following us!