At its annual I/O developer event on Tuesday, Google revealed several planned upgrades to Google Assistant, further widening the gap with Apple's Siri.
A key addition is "continued conversation," coming this summer, which will let owners of Google Home speakers ask follow-up questions for 8 seconds without repeating "OK/Hey Google." For people concerned about privacy the technology will be purely optional, toggled through the Google Home app. Microphones will also shut off automatically if someone ends a conversation with "thank you."
While continued conversation will initially be limited to Home devices, something coming to both phones and speakers is "multiple actions," the ability to ask two question in a single sentence. In fact the feature is limited to two commands, but in tandem with continued conversation should let people accomplish several actions much faster.
Another planned addition for Home speakers is "Pretty Please." When on, this will provide positive feedback for users who say "please" and "thank you," even nudging them to say "please" if they simply bark an order. It can only be enabled per voice profile, and won't prompt for words after every question in a continued conversation.
The technology is aimed at parents concerned that smartspeakers may be teaching bad manners. It will launch for Home and third-party smartspeakers starting in the summer.
Siri so far lacks any equivalent of the above technologies. Apple's voice assistant has often been criticized as lagging behind Google Assistant and Amazon Alexa, particularly on the HomePod, where it's more restricted than on iPhones.
Part of this is attributable to Apple's strict privacy stance, limiting the amount of data Siri can work from, but the company has been ramping up its investments in AI and machine learning to compensate where possible. Some of the fruits of this labor could be revealed at this June's Worldwide Developers Conference.
48 Comments
A lot of impressive things announced. Your move Apple.
More of Google just copying Amazon! /s
The “continued conversations” seems like an easy thing to implement and makes a lot of sense. I also like being able to make separate requests in the same sentence (although “turn on the popcorn maker” is probably not a great example, I imagine one would have to put popcorn into the popcorn maker prior to turning it on and at that point it would likely be easier to turn it on by hand right then. At least, that’s how it works for me...)
Have we seen any changes to Siri that can be directly attributed to any of those recent (as in the last couple of years) acquisitions we’ve heard about, like VocalIQ, etc.?