Google used its I/O 2017 developer event to unveil updates to its Assistant service and provide more details of the next version of its Android platform – as it revealed there are now more than 2 billion monthly active Android devices worldwide.
Sundar Pichai, CEO (pictured), reiterated his position the world is moving from “mobile-first” to “AI-first”, which is: “forcing us to reimagine our products for a world that allows a more natural, seamless way of interacting with technology”.
“Think about Google Search: it was built on our ability to understand text in webpages. But now, thanks to advances in deep learning, we’re able to make images, photos and videos useful to people in a way they simply haven’t been before,” he wrote in a blog post supporting the event.
“Your camera can ‘see’; you can speak to your phone and get answers back – speech and vision are becoming as important to computing as the keyboard or multi-touch screens.”
Google said almost 70 per cent of requests to Google Assistant are expressed in natural language, rather than using keywords as in typed search, and “many requests are follow-ups that continue an ongoing conversation”.
With much of the attention in 2016 focused on voice access to AI-enabled services, this year Google talked up Lens – which it said is “a new way for computers to ‘see’”.
Initially working with Assistant and Photos, the technology can: “understand what you are looking at and help you take information based on that information”, for example providing restaurant reviews or recognising Wi-Fi passwords.
Lens capabilities will also be added to other products.
Assistant and Home updates
Google said Assistant will become more proactive, delivering useful information to users when appropriate – for example providing a traffic update related to a calendar entry.
Also added was Assistant compatability for Apple’s iOS. Backing the Apple platform is not surprising – Google generally makes its services available on the rival platform – , but it will provide a strong competitor for Apple’s own Siri, although without the same degree of platform integration as is possible with Android.
In addition to broadening device support, Google is adding more Assistant languages: Brazilian Portuguese, French, German and Japanese are coming soon, with Italian, Korean and Spanish due by end-2017.
One update is decidedly retro – the ability to type interactions. This was given as a way to use the service where voice input may not be appropriate, for example when discretion is needed in public places.
There are now more than 70 smart home partners offering Assistant across Google Home and Android phones, including Nest, Honeywell, Logitech and LG.
New features have also been added to Google Home, the company’s connected speaker product.
“In the coming months” it will gain hands-free calling, with the ability to connect to mobile phones or landines in the US and Canada for free (availability of the feature for other markets was not detailed).
Its entertainment services have been updated, with the addition of the free version of Spotify (premium was already available), Soundcloud and Deezer. Bluetooth support is also being added, to enable audio to be played from iOS and Android devices.
Additional streaming video partners are on the way.
Following its recent launch in the UK, other markets are being lined-up for Google Home, including Australia, Canada, France, Germany and Japan.
The company said the platform will “bring more fluid experiences”, as well as offering improvements to “vitals” such as battery life and security.
Among the updates demonstrated were picture-in-picture, to enable two tasks to be completed simultaneously, and smart text selection.
Also detailed was what Google described as an effort to: “get computing into the hands of more people by creating a great smartphone experience on all Android devices with 1GB or less of memory”.
Called Android Go, it is apparently: “designed with features relevant for people who have limited data connectivity and speak multiple languages”. It was seen by observers as the follow-up to 2014’s Android One initiative, which targeted non-smartphone owners in emerging markets.
Android Go has three key components: a version of Android optimised for resource-constrained devices; Google apps, which have been designed to use less memory, storage space and mobile data; and a version of the Play app store which highlights apps “specifically designed for the next billion users coming online, while still offering the entire app catalogue”.
The three will be offered together from 2018.
Google showcased Play Protect, a suite of security services for Android which is “built into every device with Google Play”. It is always updating, and automatically takes actions to protect devices and data, and a “find my device” (with remote lock and wipe) offering is part of the package.
VR and AR
Google also provided an update on its Daydream VR activities.
The big news was the creation of standalone VR headsets running the Daydream platform – it previously used smartphones to power low-cost headsets. Set for launch later this year, HTC – something of a VR pioneer through its Vive business, and a long-term Google partner – and Lenovo are readying products.
Qualcomm is preparing a standalone headset reference design.
More smartphones are being brought to market with VR capability. Samsung’s Galaxy S8 and Galaxy S8 Plus will be Daydream-enabled via a software update (Samsung is another company which already runs its own VR efforts), while LG will offer Daydream with its next flagship.
Google’s Tango augmented reality (AR) platform is being used to power a service called Visual Positioning System, which enables devices to quickly and accurately understand indoor locations: “While GPS is great for getting you to the storefront, with VPS your device can direct you right to the item you’re looking for once inside,” the company said.
The company offered retail and museums as potential applications for the service.
Meanwhile Google said after launching Google Photos in 2015, the app now has more than 500 million users, backing up more than 1.2 billion photos and videos each day. It used I/O to unveil three new features.
Suggested Sharing uses machine learning technology to identify pictures which could be shared with others – for example other people in a photograph. This will encourage users to share, so (for example) pictures taken of an event by multiple people will be available to all the attendees.
Shared Libraries will enable users to automatically share pictures with one person, giving access to a full library, photos of specific people, or from a certain date onward.
And Photo Books is a service enabling users to create physical albums by selecting images from Google Photos (initially available in the US, priced at $9.99 for a soft cover and $19.99 for a hardcover book). The app removes duplicates and poor quality images, and in time Google will auto-suggest photo books.