from recent reading for those who like to consider the dots that do exist....


https://dev.smt.docomo.ne.jp/?p=common_page&p_name=sebastien_teaser



To everythingI would like to provide interactive AI service
Through natural dialogue, you can talk, Provide content, operate devices
As if everything has a will, you can dialogue. We will provide such a new experience.


the important thing to note is how extensive the Voice platform is in concept and the the pictograms clearly have Smartphones as one part of it.....



Service overview


Service providers and manufacturers provide functions that connect various services and devices by using the foundation of docomo.
We will provide environments in which AI can be introduced to everything around us and will enable discovery and suggestion of new lifestyles.


To everything around us Can carry AI

Two agents

The main agent listens to the request first for the audio request.

This is the equivalent of the wake up word we are doing already...
The main agent interprets fluctuating information specific to voice input and judges whether it is a task that can be handled by oneself.
If the content can not be solved by yourself, it will be an expert agent that provides the right service for the request
By smoothly changing, we resolve the request without stress.

So a tiered approach of local and cloud, same as we are already doing for some ecosystems...


They consistently have Smartphones as a concept..
they have to decide to get rid of push to talk?

The Japan reference board could do well for us for this whole initiative?

KT and China Mobile could be the next ones to do the same?