Implement SSML parsing at SpeechSynthesisUtterance
Categories
(Core :: Web Speech, task)
Tracking
()
People
(Reporter: guest271314, Unassigned)
Details
Reporter | ||
Comment 1•7 years ago
|
||
Comment 2•6 years ago
|
||
Reporter | ||
Comment 3•6 years ago
|
||
#2 Chromium is now performing some SSML tests though as far as am able to gather - at extension code - not related to Web Speech API, at least from what can gather by the names of the files and lack of explicitly stating the tests and the language in the comment
This change doesn't integrate the ssml parser; this will be done in a follow-up change to this one.
https://bugs.chromium.org/p/chromium/issues/detail?id=88072#c58; https://bugs.chromium.org/p/chromium/issues/detail?id=88072#c59; https://bugs.chromium.org/p/chromium/issues/detail?id=88072#c60. #61 was deleted due to again referencing that SSML parsing is possible given the effort to do so (a start using JavaScript https://github.com/guest271314/SpeechSynthesisSSMLParser).
From a front-end perspective, SSML parsing should not be abstracted to code that is not reachable/modifiable by the developer/user without undue restrictions as to how the markup is parsed to audio output. One example of the varying interpretations of the SSML specification, and implementation are
<p>
and
<s>
elements, where when tests are run on SSML third-party web services, there are different implementations. If recollect correctly, *oogle "Home' service, *mazon "*lexa", *BM *luemix inserted pause or did not insert pause arbitrarily as the language of the specification does not precisely specify what should NOT occur (https://github.com/guest271314/SpeechSynthesisSSMLParser/blob/master/SpeechSynthesisSSMLParserTest.html#L258).
Updated•6 years ago
|
Updated•2 years ago
|
Description
•