enhancedAudio clarification

[Matt Garrish]

This name seems complicated to me, as Audible, for example, call their file format for audio books Enhanced Audio. DAISY books with synchronized/structured audio are also referred to sometimes as enhanced audio books. There are a number of ways the name can be interpreted depending on your familiarity with audio, but human speech in the foreground isn’t the most obvious.

 

I see in the notes that this came from discussions around WCAG 1.4, which is about audio control of foreground/background sounds (e.g., mood music not interfering with TTS). Why not a sound control value then for this other need?

 

The definition makes it sound like an audio clarity indicator only for human speech, which is maybe too precise a meaning. I don’t have a problem with indicating the quality, but enhancedAudio doesn’t feel like the right name.

 

Matt

[Liddy Nevile]

my position on this is that auditory content is fixed or enhanced. If
there is the facility to remove the background sound eg, that would be
not just an enhancement but a refined one - and other refinements
might exist - change of voice? ??? but maybe it'd be better if it was
called adaptableAudio ?

Does conflict with names or labels of metadata really matter when we
can have machine-readable IDs? and in different languages the words
might not clash?


Liddy

> --
> You received this message because you are subscribed to the Google
> Groups "Accessibility Metadata Project" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to a11y-metadata-pr...@googlegroups.com.
> To post to this group, send email to a11y-metad...@googlegroups.com
> .
> For more options, visit https://groups.google.com/groups/opt_out.

[Matt Garrish]

I'd assumed enhanced was synchronized or structured audio until I read the
definition more closely, and I'm sure others would from the ebook world
would do the same. We can't avoid every possible conflict, but the more
precise the values are the better.

But I think I'm still not quite grasping the concept, either, so my
apologies on that. I'd thought this was a means of identifying multi-channel
audio so the user could potentially disable the background audio. It sounds
from your description that it's actually indicating that alternate audio
files are included, correct? So you could have a noise-reduced version, one
spoken in another voice or dialect, etc.? If so, I'd prefer adaptableAudio
to convey that sense that there is choice.

And I agree that numeric IDs can be better for internationalization, but
those systems are difficult to make workable except in controlled systems
like ONIX. Even if a search engine were to translate numeric meanings, they
exact a toll on content producers who can't remember what number means what,
can't easily verify if their metadata is correct without looking up numbers,
can't be sure how to extend, run into future conflicts, etc. It's also
somewhat antithetical to the concept of semantic data for search engine
optimization that schema.org espouses. Numeric values would imply metadata
is never visible and is not text content of the page.

Matt

[Charles Myers]

Good to have you back editing, Matt.

None of us on the calls knew that this term was used in multiple other places and was, in a sense, overloaded.

The history on this is pretty simple:

1) Charles brought this up as an example of an kind of accessibilityFeature (nee mediaFeature). We had no idea what to do with it, though

2) Then I happened to be reading the WCAG guidelines (1.4) for the largeText/Print and resizeText and found the basis for this in 1.4.7 http://www.w3.org/TR/UNDERSTANDING-WCAG20/visual-audio-contrast-noaudio.html

From there, there were a few candidate names

audioContrast

highContrastAudio

enhancedAudio

Since it was not just about contrast, but also about eliminating the background completely, the contrast names did not work.  That left us with enhanced, but with the problems you point out.

With 20/20 hindsight, we also could have called this "lowOrNoBackgroundAudio"  Not the sexiest, but it stays close to the meaning of the specification.  All three of the values we describe make even more sense this way.

[Matt Garrish]

Ah, okay, each email clues me in a little further! (I guess my thoughts on
adaptable weren't quite right, either.)

The highContrastAudio option jumps out as the one that best reflects the
distinction between foreground speech and background noise in that SC.
Contrast is the key concept, and being enhanced it was makes it more
accessible, so enhancedAudio alone lost that meaning for me and sent me off
in other directions.

Matt

-----Original Message-----
From: Charles Myers
Sent: Thursday, December 05, 2013 8:57 PM
To: Matt Garrish
Cc: <a11y-metad...@googlegroups.com>
Subject: Re: [a11y-metadata-project] enhancedAudio clarification

<matt.g...@bell.net<mailto:matt.g...@bell.net>> wrote:

This name seems complicated to me, as Audible, for example, call their file
format for audio books Enhanced Audio. DAISY books with
synchronized/structured audio are also referred to sometimes as enhanced
audio books. There are a number of ways the name can be interpreted
depending on your familiarity with audio, but human speech in the foreground
isn’t the most obvious.

I see in the notes that this came from discussions around WCAG 1.4, which is
about audio control of foreground/background sounds (e.g., mood music not
interfering with TTS). Why not a sound control value then for this other
need?

The definition makes it sound like an audio clarity indicator only for human
speech, which is maybe too precise a meaning. I don’t have a problem with
indicating the quality, but enhancedAudio doesn’t feel like the right name.

Matt

--
You received this message because you are subscribed to the Google Groups
"Accessibility Metadata Project" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to

a11y-metadata-pr...@googlegroups.com<mailto:a11y-metadata-pr...@googlegroups.com>.

To post to this group, send email to

a11y-metad...@googlegroups.com<mailto:a11y-metad...@googlegroups.com>.