<scribe> scribe:ddahl

architecture document https://w3c.github.io/voiceinteraction/voice%20interaction%20drafts/paArchitecture-1-1.htm

debbie: review changes in the introduction and additional standards

dirk: specialized assistants in Amazon ecosystem need a trigger word
... should describe how these are triggered somewhere in the document

jon: expect that 25% of enterprises will have a voice assistant by 2025
... can share that data
... they have to be interoperable
... David added IPA in ASR and ISO-2476-12 for DIT++
... TEI (text encoding initiative)

debbie: MRCP (Media Resources Control Protocol)

jon: Open Agent Architecture for provider selection service
... in accounts and authentication (OATH) and Open ID
... in Core NLU -- abstract meaning representation
... can send links
... also ARPA NGRAM
... which is the same as the SRI-LM format

debbie: should we add a column for the responsible standards body?
... we don't need to go into that
... may not be worth the effort
... some may be obsolete or not adopted

dirk: should add a disclaimer at the beginning

debbie: some may be standards that we would adopt and others might just be informative

dirk: could add dialog modeling language from the other community group

debbie: not saying anything about quality or usefullness
... of these standards and some are de facto standards or proposals
... also, what about audio formats and codecs?
... or is that too low-level (ogg vorbis, speex)

dirk: this could be endless
... this could be important if you need to forward audio

debbie: we won't talk about audio codecs, but they need to be agreed upon
... could we use whatever the Audio Working Group uses?

dirk: doesn't see that in Web Audio document

https://www.w3.org/TR/webaudio

<scribe> ACTION: debbie to investigate web audio to see if they say anything

debbie: agnostic to low level protocols, e.g. web sockets and http
... but then it won't work if the service doesn't understand the protocol

dirk: WebRTC
... what about IPA Service?

debbie: there is an open source smart speaker called Mycroft, is there an open way to configure it?

dirk: that's more of an IPA client
... the closest might be a web service

debbie: would REST be an example?

dirk: what about "core intent set"?

debbie: DAMSL is relevant but more abstract than core intents
... core intents are more specific

dirk: a core intent could be "help"

debbie: some core intents are defined in VoiceXML ("builtins") -- help, nomatch, noinput and error

<scribe> ACTION: debbie to start editing in these changes, review next call

debbie: should publish when these edits and caveats are added
... maybe the beginning of the year would be a good goal

<scribe> ACTION: debbie to look into host key problem

This is scribe.perl Revision of Date 
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Present: debbie jon dirk
Found Scribe: ddahl
Inferring ScribeNick: ddahl

WARNING: No "Topic:" lines found.


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: debbie

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report


WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)