W3C

– DRAFT –
Spoken Pronunciation Task Force Teleconference

03 February 2021

Attendees

Present
becky, Dee, Irfan, IrfanA, janina, JF, NeilS, paul_grenier, Roy, shadi, SteveNoble, SteveNoble_, tom_babinszki
Regrets
-
Chair
Paul Grenier & Irfan Ali
Scribe
JF

Meeting minutes

Agenda Review & Announcements

IA: any announcments?

JS: nothing much here. noted that the doc is getting ready for publication.

JS: Charter issues are with a different group at APA

IA: we're doing well. On track re: timelines

Technical Approach document https://w3c.github.io/pronunciation/technical-approach/

IA: some changes are bing made - but asking for wider review

PG: I have some proposed changes local - will need to figure out how to merge

IA: is there a schema elsewhere that we can just link to? Rather than adding the schema into the document?

MH: not sure what W3C policy is there

JS: that's a M. Cooper question for later today

MH: we should be able to point to the external doc

JS: that allows the two documents to advance independantly

NS: but that maigh t have a negative impact too

MH: we need to define this

JS: can add this as an appendix

MH: that's the approach I am taking now

NS: why do we need a schema for the singular approach?

PG: not so much a schema but a list of elements. Anything that could have more than one attribute - we need something of a namespaciong mechanism

MH: tried to get that as specific as possible for the multi-attribute model

IA: like Mark's idea - are we good with it?

IA: we need to publish a sample page for testing/demo purposes

NS: we need to ensure parity to SSML spec. Now that everything is really spelled out, have concern there may be a divergence

we might get o9ut of sync with SSML
… so maybe we should just do the reference

MH: the issue is that we need to be explicit about the mapping

NS: we can just do a reference...

MH: started to do that in a few places (say-as): some of the attribute values are defined in a separate W3C note
… so every instance needs a hard link to the appropriate spec or note
… but there shold be a source "readable" version in our spec

NS: think it's important to reference specific sections

MH: stopping editting for now - need to leave it to settle
… let Paul add his edits, then ask M. Cooper or Roy to ensure prooper W3C formatting

PG: most of my edits are the prose section - using Plain Language tools to try and achieve more readable specs (ref: WCAG 3 et. al)
… don't anticipate needing to touch any of the code examples

MH: re: sample pages - we need to extract out the code examples for both approaches into a single sample page for testing purposes (include the RAven example)

IA: is github the best place to host those pages?

MH: best place to maintain them , yes

PG: ...exporting the same files... (???)

[discussion about exporting/Alexa/audio files]

MH: will take that as an action item (Irfan and mark)

IA: as far as wide review - who/how?

JS: APA calls for the review, and does the publishing. As to who we reach out to - we need to discuss
… Shawn Henry has some resources about messaging etc. Will share URL on list shortly
… but we need to start filling out the forms - who do we want to reach, how (twitter, blog posts, etc.)
… Judy will be quite interested about this. She will anticipate we've already spoken to Shawn

MH: recalls that we need to document implementations

We have one example already (TextHelp), plus some experimental PoC's
… so collecting implementations is importatn

IA: agree - very important point

<janina> Messaging guidance is here:

<janina> https://www.w3.org/wiki/WAI/announcements

IA: is this something that we can template?

MH: thinks we can do by reference

MH: we need to be practical about this

<IrfanA> JF: the policy at w3c is any specific advances to recommendation, you need to ave two implementation.

JS: we must have 2 independant examples of each (RFC 2119) MUST requirement

<IrfanA> janina: little correction.. two out of 5

JF: this is a must have

MH: propose a wiki page to collect implementations at this time

key point for this in timeline is CR (for examples)

IA: do we need APA support for request for implementations?

JS: no - can go ahead and just "do it"

IA: Shawn is aware of this work - will reach out to her

JS: I posted the URL

IA: sounds like we have a plan. We need to publish by end of Feb - believe we are on schedule

IA: will joind APA call to ask MCooper aqbout eh Schema questions

Other Business

JS: nothing from me

<IrfanA> JF: created a wiki page and it is in my list of things to do to identify the venues to post about the development

https://github.com/w3c/pronunciation/wiki/Socialization

JF: not the size so much as the attendees

Minutes manually created (not a transcript), formatted by scribe.perl version 127 (Wed Dec 30 17:39:58 2020 UTC).

Diagnostics

Maybe present: IA, JS, MH, NS, PG