Draft Decision Letter

Based on the conversations following TPAC, we, the Pronunciation Task Force
(TF), decided to pursue the "single-attribute" approach for Specification
for Spoken Presentation in HTML <https://www.w3.org/TR/spoken-html/>.

Following that decision, the ARIA Working Group (WG) provided additional
feedback <https://www.w3.org/2021/12/01-pronunciation-minutes.html> that we
would like to address. In an effort to reach prototype implementation, we
propose to scale back the eight (8) sections of the specification
(currently section 3.1
<https://www.w3.org/TR/spoken-html/#data-ssml-attribute-properties-and-values>)
to four (4):

   - say-as <https://www.w3.org/TR/spoken-html/#say-as>
   - break <https://www.w3.org/TR/spoken-html/#break> (pausing)
   - phoneme <https://www.w3.org/TR/spoken-html/#phoneme>
   - sub <https://www.w3.org/TR/spoken-html/#sub>

Based on the gap analysis
<https://w3c.github.io/pronunciation/gap-analysis/#gap-analysis>, the TF
feels these qualities of speech represent the most value to the widest
audience and propose they be included in the first phase of prototype
development.