Topics
Published on December 8, 2016 by Google

Learn more from the documentation at our developer site: developers.google.com/actions/
Google+ Actions on Google Developers Community: g.co/actionsdev
Watch the “Actions on Google: Conversation Design Tips” video: goo.gl/rBilJL

Thanks to Sci-Fi, many of us have pictured a future where we can carry on a full conversation with the objects around us. But will we ever be able to talk to things like we talk to people? Let’s face it. Right now, most voice user interfaces fall a little short of being, well, conversational. In this video, we take a look at what happens when two humans have a conversation. Once we can understand that, let’s figure out how we can apply it to modern interfaces.

Want to learn more and be part of a conversational future? We can help you do that with Actions on Google. Visit our design resources page for lots of tips to get you started at: developers.google.com/actions/

Subscribe to the Google Developers Channel: goo.gl/mQyv5L

Leave a Reply

7 Comments on "Actions on Google: Conversation: The New UI"

Notify of
avatar

Ayaz Alam
Guest
Ayaz Alam
4 months 14 days ago

Loved the animation!

Mert Kırımgeri
Guest
Mert Kırımgeri
4 months 15 days ago

bullshit.We enter the answers.Where is the actual learning ? I did that before with a bunch of data contain answers and just a levenshtein function.

Nicolás Parada
Guest
Nicolás Parada
4 months 16 days ago

Very exciting

Eni Ien
Guest
Eni Ien
4 months 16 days ago

Does Google assistant uses the same architecture as Apple's Siri?

Balistic_ Penguin
Guest
Balistic_ Penguin
4 months 16 days ago

For text-based conversation, it would be great to partner up with an already popular messaging app (like kik or discord) to help neural networks to learn off of a bunch of regular conversations.

Mohammed Shaikh
Guest
Mohammed Shaikh
4 months 16 days ago

which type of animation is done in this video..?

Gabriel Kwiecinski Antunes
Guest
Gabriel Kwiecinski Antunes
4 months 16 days ago

I've dreamed about university students searching for official documents and printing them with autocompleted fields when possible (name, id, registration numbers, etc…) by talking to their phones or the computers at the university library. That seems like a reasonable first step. It's great to know that you guys are managing the voice recognition heavy lifting and I'm sure the abstractions you developed for Actions will help many good ideas considered too hard for so long to finally bloom. Thanks. For real.

wpDiscuz