3 ms·
I did a whole write up about my homegrown attempt to do this last summer. That blog post from Google is more technical but eerily similar. https://philandrews.i
by isthispermanent 8y ago
I did a whole write up about my homegrown attempt to do this last summer. That blog post from Google is more technical but eerily similar. https://philandrews.io/post/20000-phone-calls-later-why-siri-cant-make-your-dentist-appointment https://philandrews.io/post/20000-phone-calls-later-why-siri...
In short, I don't have a lot of faith in this being plausible yet. The major reason is the phone lines. Phone call quality is not solid enough to ensure a accuracy rate high enough to roll this out as a production service. There are others, like legal considerations, but if I had to pick one that would be it.
There's a reason why this rose to number one HN, people would clamor for it. To think that Apple and Google haven't been thinking the same thing is short sited.
- zerostar07 8y agoIt was the most impressive thing in the keynote for sure, but given the current state of AI we should have no reason to believe that this would work well enough to be employed at scale. It also feels weird that it is trying to trick the callee that it is a human. I believe even if they used a mechanical voice, the system could work, because businesses would learn to recognize the "google call" and respond to it the way they respond to any voice-enabled call center. It seems to me it is more of an (impressive) PR thing than something that will get actual use.
- isthispermanent 8y agoI agree. There's a whole world where I imagine businesses with stickers like the Visa accepted here stickers, except something like "Siri accepted here" or "Siri works here", whatever the wording. That way you would know that you could use Siri to interact with that business on your behalf. That takes away part of the resistance to phone bots by people that answer phones, them expecting those calls.