W3C home > Mailing lists > Public > public-html-a11y@w3.org > May 2011

Re: Action item. definition and use of Clean audio in European television

From: Geoff Freed <geoff_freed@wgbh.org>
Date: Thu, 26 May 2011 05:32:26 -0400
To: "public-html-a11y@w3.org" <public-html-a11y@w3.org>, David Singer <singer@apple.com>, Silvia Pfeiffer <silviapfeiffer1@gmail.com>
Message-ID: <CA03966A.69BB%geoff_freed@wgbh.org>

Hi, everyone:

I've been following this discussion from the sidelines, and at the risk of confusing things I offer the following bit of informal research.  I've spoken with broadcast colleagues here at WGBH and also at the BBC to see what "clean audio" is understood to mean in the US and in Europe:

-- In the US, "clean audio" means, from a technical perspective, high-quality audio that is free of artifacts or other sonic garbage.
-- In Europe, "clean audio" refers to soundtracks with enhanced speech intelligibility.  It seems that confusion arises over whether it refers to soundtracks with *all* music and sound effects removed, soundtracks on which non-speech elements have been merely reduced or just the center channel of a 5.1 mix.  I also received the following comment:  "We [the BBC] don't like clean audio as a term, because it presumes the solution to improved-intelligibility audio is to lower/remove music and [effects], whereas we've found it to be much more complex than that."

Geoff
WGBH/NCAM


On 5/24/11 11:26 PM, "David Singer" <singer@apple.com> wrote:

As I say, I don't mind if it says "foreground speech" or something, and the documentation says it's the foreground sound that can be isolated for those who need it, and it's typically speech

On May 24, 2011, at 15:31 , Silvia Pfeiffer wrote:

>
> On 25/05/2011, at 5:35 AM, David Singer <singer@apple.com> wrote:
>
>>
>> On May 23, 2011, at 21:39 , Silvia Pfeiffer wrote:
>>
>>>
>>> Not quite when you look at those that talk about "clean audio" in the
>>> way that we do. There, only "foreground dialogue and speech" is
>>> explicitly mentioned, e.g. http://www.guidogybels.eu/cap3.html. Note
>>> that the overall effect of the presentation may be to receive "clean
>>> audio" in the way that you describe, but the track's content is not
>>> clean audio, but only "foreground dialogue and speech".
>>>
>>>
>>>>>
>>>>> ETSI actually call a channel that contains speech-only for "clean
>>>>> audio" purposes a "hearing impaired" channel. At least this describes
>>>>> what the channel is being used for. "speech" would describe what it
>>>>> contains. "clean audio" give a false indication of better sound
>>>>> quality. I would be ok with "foreground sound", too, but I would be
>>>>> very unhappy about the term "clean audio".
>>>>
>>>>
>>>> How do you feel about 'contrast enhanced' or 'high contrast' or 'low distraction'?
>>>>
>>>> But...as you say...I'd like to use industry terms even if, to our ear, they are not ... very clean. :-)
>>>
>>> I think you believe that other sounds than speech and dialogue are
>>> also enhanced/extracted for "clean audio". That is explicitly not the
>>> case IIUC.
>>
>>
>> I think we can use an industry term, such as 'clean audio' or 'clean speech', and make it clear in the documentation that this is the 'primary program material, without extraneous or aesthetic background sounds'.
>>
>> I imagine that in a program about whale-song, with schlocky new-age poetry being read in the background for 'effect', the 'clean' audio would be the whale-song without the poetry, for example.
>
> I doubt that actually. What users want is to manipulate the loudness of the speech separately from the rest because it's in human mature to care about speech most. The thing is: it also allows you to fade out the speech giving cleaner access to the rest of the sound scene. So, the effect is "clean audio".
>
>
>> But given decent documentation, the author could work that out, even if the label says 'speech'.
>
> If you prefer, we can call it "foreground sound" - that's more semantic and less confusing than "clean". I was going for "speech" since that is 99% of the content and people intuitively understand the idea of turning up the "speech" track (in contrast to turning up the "clean" track).
>
> Silvia.
>>

David Singer
Multimedia and Software Standards, Apple Inc.
Received on Thursday, 26 May 2011 09:32:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 04:42:38 GMT