[voiceinteraction] minutes January 12, 2022

https://www.w3.org/2022/01/12-voiceinteraction-minutes.html
and below as text

   [1]W3C

      [1] https://www.w3.org/

                             - DRAFT -
                           Voice Interaction

12 January 2022

   [2]IRC log.

      [2] https://www.w3.org/2022/01/12-voiceinteraction-irc

Attendees

   Present
          bev, Debbie, dirk, Jim, jon

   Regrets
          -

   Chair
          Debbie

   Scribe
          ddahl

Contents

    1. [3]2022 plans
    2. [4]architecture
    3. [5]JSON semantic representation
    4. [6]Summary of action items

Meeting minutes

  2022 plans

   debbie: let's talk about interfaces document

   debbie: just getting started

   debbie: talked about priorities for communication

   dirk: shows interfaces diagram

   debbie: client input and client response

   debbie: client input will mainly be speech + metadata

   dirk: would EMMA be suitable?
   . properties are in section 4.1

   debbie: output will also be audio in general

   dirk: could also be text

   jim: do we need to list other kinds of multimodal data?
   . it seems kind of vague

   dirk: yes, didn't define this precisely

   jim: more types of metadata?

   debbie: let's not detail these at this point

   jim: what do you mean by multimodal input?

   dirk: pen, text,

   jim: should add some examples of multimodal inputs

   jim: are these categories or data items?

   jim: could distinguish categories from data items
   . metadata and multimodal are categories

   dirk: add another column for category

   <Bev> Hello

   debbie: do we agree that these are the priorities

   jim: yes

   debbie: the client response is primarily audio

   dirk: should we add a request id?

   debbie: yes

   jim: identifying major elements of client input and response
   . the other communication paths are ASR input and semantic
   input
   . these are the priorities
   . will build tables for these

   debbie: can build EMMA examples

   jim: should we build EMMA examples or finish building tables

   debbie: let's do both at the same time

   ACTION: dirk to fill in tables

   ACTION: debbie to build EMMA examples

   debbie: we should publish an update then

  architecture

   debbie: should we update the architecture?

   jim: we should do some publicity for architecture

   jon: has reached out to people in the voice world, for example
   Project Voice in Chattanooga

   debbie: could write about in standards

   dirk: will check about Conversational Interaction

   jim: is there a place in Europe?

   ACTION: jim to introduce Dirk to Marten

   jon: voice lunch is focused on conversational design

   jon: there are places to go to push this work into the public
   eye

   jim: what about a blog?

   jon: we can publish on LinkedIn, but the problem is getting
   people to read it
   . same thing about podcasts

   debbie: should we plan to update the architecture?

   dirk: probably not now

  JSON semantic representation

   debbie: is it still important to have a JSON version of EMMA?

   dirk: it would be beneficial, because JSON is preferred

   debbie: review JROSI

   jim: should that be a different community group?

   debbie: would there be some participation?

   jim: organize a community group about semantics

   debbie: people like to start something new

   jim: what about in the W3C?

   debbie: new W3C workshop

   jim: presentation at W3C plenary?

Summary of action items

    1. [7]dirk to fill in tables
    2. [8]debbie to build EMMA examples
    3. [9]jim to introduce Dirk to Marten

Received on Wednesday, 12 January 2022 19:37:45 UTC