Reference Speech Synthesis Markup Language (SSML) Version 1.1 at HTML Living Standard

[guest271314]

SSML Speech Synthesis Markup Language (SSML) Version 1.1 is presently not implemented at either Chromium/Chrome or Firefox when a valid SSML ("document") is passed as first argument to SpeechSynthesisUtterance.

Following the suggestion within https://github.com/whatwg/html/issues/3282 this issue specifically seeks to

1) Reference Speech Synthesis Markup Language (SSML) Version 1.1 here at WHATWG HTML Living Standard

1) create and maintain an open source code base which implements SSML parsing as described at the specification which can then be implemented at modern browsers at calls to SpeechSynthesisUtterance()

[domenic]

Why would we reference it? It's not part of or related to the HTML Standard. There are many specs on the web platform but we only reference the ones that HTML depends on.

As for 2, that sounds like an interesting effort, but completely unrelated to the HTML Standard as well.

I'll close this, as there's nothing actionable for the HTML Standard as far as I understand, but I am happy to reopen if I misunderstood.

[guest271314]

@domenic Not sure how to go about what am attempting to convey. Am far from yet fully understanding the boundaries and politics of specifications at W3C, WHATWG, implementation code at browsers, RFC, etc. From perspective here, work has essentially stopped relevant to the Web Speech API specification - and implementation. Though there is more work to be done to even realize the implementation described at the linked specification.

This subject matter relates to accessibility technologies rendering at markup - SSML. WHATWG is quite active and progressive in the methodology of approaching subject matter. The actionable aspect would be seeing the existing specification through to the fruition of implemented code at browsers - that is, so that there would be consistent application of the specification using the same code composed by viewers of the referenced specification here at WHATWG.

[domenic]

In general you'd be best finding the author of the specification and contacting them if you want to improve that specification. If you want that specification to be implemented in browsers, you should open bugs on browsers asking them to. Does that help?

[guest271314]

@domenic Yes, that helps. Have attempted to contact the "OWNERS" of Speech at Blink, as evidenced at the bug link, have not tried to contact Mozilla component owners yet. Previously contacted the author of the specification at https://lists.w3.org/Archives/Public/public-speech-api/2017Jul/0004.html, https://lists.w3.org/Archives/Public/public-speech-api/2017Jul/0005.html for a feature request.

Perhaps the issue is lack of currently being equipped to compose the C++ source code implemented at browsers, and it is simply the signal to learn C++ and other relevant languages and build processes to compose the necessary code.

[guest271314]

@domenic FWIW https://bugs.chromium.org/p/chromium/issues/detail?id=795371, https://bugzilla.mozilla.org/show_bug.cgi?id=1425523

[guest271314]

@domenic Evidence that SSML is related to accessibility within HTML https://github.com/mhakkinen/SSMLinHTMLproposal, https://github.com/mhakkinen/SSML-issues/blob/master/overview.md

[domenic]

I'm not really interested in this topic, and I'm not sure the other 231 watchers of whatwg/html are either, so I'd appreciate if you stopped pinging this thread.

[guest271314]

@domenic Oh, then this is not really an open source mechanism for improving the Web, correct? Only open as to contributors for what the "Owner"'s consider relevant or are "interested" in? You may be astute, though you cannot get into 231 individuals' minds to determine what their individual or collective interests are. Have no issue being banned or censored from any board on the web simply for the way communicate; certainly not for attempting to get SSML technologies implemented. Being as though you and your co-owners have and do contribute to the Web, will step aside, here, without rancor. Cheers.

[annevk]

This repository is meant for improving the HTML Standard. I tend to agree with the assessment that there does not seem to be overlap between what HTML defines and SSML defines. It seems a discussion around SSML would be more appropriate on https://lists.w3.org/Archives/Public/www-voice/, given that's what the standard itself encourages.

[guest271314]

@annevk Again, the work that is performed here at HTML Living Standard is studious and certainly moves towards stability and improvement of the Web. However, when "Owner"s are able to pick and choose what topics are brought into the fold based on what they are "interested" in, or only upon their own assessment of a causal or substantive relationship between HTML and related subject matter, that is no longer open source contribution, but top-down bureaucracy.

The comment by @domenic could have been far more neutral in the assessment of whether or not SSML is related to HTML or other open Issues at HTML Standard without the statements "I'm not really interested in this topic" and "I'm not sure the other 231 watchers of whatwg/html are either"; the former being an individual opinion devoid of a technical reason therefor; the latter assuming the responsibility of speaking for what is in other individuals' minds - which is not possible.

You folks do a good job here. Am attempting to depart from this board without rancor; though you cannot feasibly massage the rationale for not addressing the relationship between accessibility technologies (a current open Issue here at HTML) and SSML under the premise of "I'm not really interested in this topic", which is not a technical evaluation, but an individual statement of interest - by an "Owner" of the repo - but an individual statement of interests nonetheless, not having anything to do with improvement of the Web.

[annevk]

That is fair, but @domenic said more than just that, in particular his substantive feedback came much earlier:

Why would we reference it? It's not part of or related to the HTML Standard. There are many specs on the web platform but we only reference the ones that HTML depends on.

[domenic]

Right. @guest271314, I certainly didn't intend to make you feel unwelcome or that you had to depart from this repository; I apologize for that. But I do think you need to realize that, like any repository on GitHub, this one is focused around developing a specific piece of software (in our case, the HTML Standard), and that coming to promote other pieces of software you like to the people working on it is not very appropriate.

You seem to have done a good job engaging with the relevant folks for your SSML work, and I'd encourage you to continue doing that. I do think, though, that you should not be pinging me directly, or the other watchers indirectly, to keep us updated on those discussions. Instead, let people interested in watching the SSML mailing lists or the mhakkinen repositories get notified, as already happens automatically.

If you're still unclear, I'd encourage reading https://whatwg.org/working-mode in more detail.

[guest271314]

@annevk Simply put, from perspective here, because the individuals whom contribute to the HTML Living Standard get things done. SSML implementation has not been achieved, though the standard has been published for over 5 years. Let's get this implemented in browsers in an open source (FOSS) manner instead using Alexa, Watson, etc.

[annevk]

I see, I think that kind of high-level discussion would be better in https://github.com/whatwg/meta. We'd like to keep the various repositories for particular standards focused on those standards.

[guest271314]

@domenic No worries. Have no issues with direct communication; prefer that form of exchange to subtle hints or innuendo. "pinged" you directly based on the fact that the open Issue concerning accessibility does have a causal relationship to SSML, as evidenced by the linked repo. Am not versed in the politics of whom does what and the demarcation lines between W3C, WHATWG, RFC, etc. Just trying to do what can to illuminate the subject matter to those of you on the Web who are actively working to improve the Web.

@annevk Will follow the links and continue in whichever manner am able to realize implementation of specifications that might help improve the Web. As previously mentioned it is probably high time to learn C++ and other necessary languages to be able to implement these specifications by self - then share the work - so as not to disturb what you, or other repos have already established as your methodology of action.