Re: Draft Comment for Discussion -- a11y-review issue 232

Thank you Janina for your thoughtful comments which I agree with.

And Hello all,

You all may remember me as I responded to your courtesy welcome email a few months ago. I have been following the proceedings of your emails since, but the 9am meetings continue to evade me, as we are still ambling towards a punctual daycare drop off routine.

Anyway, I 100% agree with Janina, that the use of non-speech sounds (auditory displays and sonification) needs acknowledgement and design consideration for all future web-audio specs.

While I have been developing auditory displays for the last decade, I constantly encounter constraints with webaudio. Over the last few years my focus has shifted to an even more unaddressed issue : the use of earcons compatible with the capacities of assistive hearing technologies.

yes ! rarely do folks think of accessibility beyond the sensory-substitution metaphor.

Without going into the ableist history of the speech vocoder's development, I will also add that a common concern raised by the hearing aid and cochlear implant listener community, is not having the ability to shift down the formants of 'Google lady' or TTS voices down to the sub-100Hz speech formant frequencies, to make it more intelligible and less annoying.

I hope I can contribute more, as I wrap my head around all the different meetings you all have scheduled over the coming weeks!

Looking forwards
Regards

SHARATH CHANDRA RAMAKRISHNAN
(M.S., Ph.D.)
Perceptual Futures Laboratory
Assistant Professor, School of Art & Design
Affiliate Assistant Professor, School of Information Sciences
University of Illinois Urbana-Champaign
scram@illinois.edu<mailto:scram@illinois.edu>

[https://webtools.illinois.edu/webservices/js/ds/signature_logog_165.png]<http://illinois.edu/>

Under the Illinois Freedom of Information Act any written communication to or from university employees regarding university business is a public record and may be subject to public disclosure.
________________________________
From: Janina Sajka <janina@rednote.net>
Sent: 28 May 2025 07:24
To: W3C WAI Accessible Platform Architectures <public-apa@w3.org>
Subject: Draft Comment for Discussion -- a11y-review issue 232

Colleagues:

I'm a bit weak on the mechanics of editing a github issue, so am posting
my proposed draft text here.

Last week I agreed to further refine our proposed comments on
a11y-review issue 232:

https://urldefense.com/v3/__https://github.com/w3c/a11y-review/issues/232__;!!DZ3fjg!_y7ve32rLFLRKmmCNI4v2G1TLe0KvkLbcPAGhlfbzy0vQTAw4tWsTDTEfxc4gpNzzHnoGbzpagZNwXpg$

I'm proposing an additional paragraph to account for last week's
teleconference discussion as denoted below:

<begin issue text>
In reviewing "how audio is rendered and interacts with other audio" in
the Audio Session API https://urldefense.com/v3/__https://www.w3.org/TR/audio-session/*abstract__;Iw!!DZ3fjg!_y7ve32rLFLRKmmCNI4v2G1TLe0KvkLbcPAGhlfbzy0vQTAw4tWsTDTEfxc4gpNzzHnoGbzpav1nZ-lP$ , APA
are concerned with implications for audio generated by TTS--which could
be AT generated audio in a user's environment. We see no mention of TTS
in the spec.

Did we miss it? Clearly any AT use of TTS would make that audio the most
critical audio to the user.  Should we be concerned?  Should the API
call out TTS as of particular sensitivity, do we then risk user privacy
concerns?

<proposed new paragraph begins>
Of course most, though not all AT is provided and managed by the OS,
obviating privacy concerns. Audio in AT is not limited to TTS however.
Also, there are additional uses of audio for accessibility reasons. One
key use is a short distinctive audio compositions used as signage to
signal a specific event to the user in a manner that communicates far
more quickly than TTS, allowing TTS to function as a confirmation. These
short audio compositions generally come as sets of files and are mapped
to common events (such as dialog popups) , and are known variously as
[https://urldefense.com/v3/__https://en.wikipedia.org/wiki/Earcon*5D(auditory__;JQ!!DZ3fjg!_y7ve32rLFLRKmmCNI4v2G1TLe0KvkLbcPAGhlfbzy0vQTAw4tWsTDTEfxc4gpNzzHnoGbzparEMCigB$  icons,</a> or more
descriptively as <q>Sonicons</q> or <q>Earcons.</q> At a minimum APA
considers that developers should be aware of this wider use of audio.
Certainly, when it's web-based AT, specific considerations would then
pertain.

BTW:  Your API publication has caused us to reconsider our
self-assessment questionnaire.  It should have exposed questions such as
these! Consequently we'll now be adding audio-related self-assessment
questions as we revamp our tooling.

--

Janina Sajka (she/her/hers)
Accessibility Consultant https://urldefense.com/v3/__https://linkedin.com/in/jsajka__;!!DZ3fjg!_y7ve32rLFLRKmmCNI4v2G1TLe0KvkLbcPAGhlfbzy0vQTAw4tWsTDTEfxc4gpNzzHnoGbzparBXMb9x$

The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
Co-Chair, Accessible Platform Architectures     https://urldefense.com/v3/__http://www.w3.org/wai/apa__;!!DZ3fjg!_y7ve32rLFLRKmmCNI4v2G1TLe0KvkLbcPAGhlfbzy0vQTAw4tWsTDTEfxc4gpNzzHnoGbzpaqZQ82NA$

Linux Foundation Fellow
https://urldefense.com/v3/__https://www.linuxfoundation.org/board-of-directors-2/__;!!DZ3fjg!_y7ve32rLFLRKmmCNI4v2G1TLe0KvkLbcPAGhlfbzy0vQTAw4tWsTDTEfxc4gpNzzHnoGbzpas3N56lX$