Gadgetory

so what you're gonna hear is the Google assistant actually calling a real salon to schedule the appointment for you let's listen depending on what service she was like what service is she looking for okay we have a 10 o'clock echo potato how may I help you hi I'm the Google assistant calling to make a reservation for a client this automated call will be recorded can i book a table for Tuesday the 12th okay cool we've always been thinking about how should the system kind of introduce itself and the way that we think about it is really having four key elements first is starting by saying what the purpose of the call is because for the business that's really the important that you're calling for a reservation or what are you calling for second is disclosing that this calls from Google third that is from an automated system so not just a human operator at Google but an automated system and then forth and states that require it we will disclose that we're recording the call great and when did you say they want to come in today at 7:00 p.m. hey let me check making those systems sound and feel more natural you have a higher success rate of of completing the call successfully for kind of the business side and the customer side and so as we kind of study that look for ways to make it natural we came upon this idea that linguists call speech disfluencies as ohms and ahhs that we all use out every day in conversation as linguists have studied this they found that those words or non words actually play a crucial role in helping the conversation to continue and so they play some really important purposes in our system to help both sides of the conversation kind of though that the other side still listening or introduced things like Corrections and clarifications hey Google look at table for two a telco gotaro on Tuesday at 7:00 we're testing we're gonna go kind of slow take it step at a time so we're starting by testing with a group of trusted testers a fairly small group and a set of partner businesses that we work with really to get feedback both from those and end-users and from the businesses and how they respond to this kind of technology all right I'll call it a book under your name and phone number and I'll update you in the next 15 minutes is that okay perfect thanks a lot of the concerns that we heard about duplex and good concerns as we announced at i/o but some of them were kind of around Wow has Google created a general AI that can have any conversation and act like a person and how will keep them know what it is and I wanted to make clear that Runa made clear that it's very far from what we've actually done what we've actually done is trained very deep models that know how to do very narrow tasks holiday hours restaurant reservations booking a haircut appointment those are really the three things that you can do so if you think duplex is like a human it's a very interesting human that can only do those three things I don't have seven but we can do eight where the system doesn't know what the right next step is and so within the calls what you'll hear is the system will try a couple times usually to kind of bring it back to what it understands so you know if you're talking system and you say well something it doesn't you know something crazy oh your Google what's the weather tomorrow what you'll hear the system do is say something like oh I'm not sure I understand but could I get a table for four on Tuesday right so kind of try to bring back if that doesn't work the system is able to signal a human operator and that human operator can either come in and just step in on the call and so what you'll hear is someone saying sorry it sounds like our automated system was having trouble completing the reservation this is this Dave on the human and you know can let's continue the conversation or alternatively they can the the automation can simply bow out and say I'm not sure we're getting through this so let me have someone call you back and then have the human call back okay we'll see Anna Tuesday thank you okay awesome thanks a lot