Re: "Protocol" requirement - Interpretation

On 12/14/2010 04:42 AM, Bjorn Bringert wrote:
> I don't have any objections to this as a requirement other than that
> it adds API complexity. It's not a requirement that I would prioritize
> highly.

I agree. I doubt API proposals will prioritize this.
Sounds more a like "version 2" requirement (whatever that means), but
sure we can keep it.


-Olli


>
> /Bjorn
>
> On Mon, Dec 13, 2010 at 7:19 PM, Young, Milan<Milan.Young@nuance.com>  wrote:
>> XmlHttpRequest may work fine, but it assumes a largely bypassed UA to handle other requirements such as session tracking.  It also assumes a SS with an HTTP server.
>>
>> So while I share your optimism on these points, I agree with Michael that we need to be careful.
>>
>> Any objections with adding this to spec even if it may later be rendered trivial by device approach?
>>
>> Thanks
>>
>>
>> -----Original Message-----
>> From: Bjorn Bringert [mailto:bringert@google.com]
>> Sent: Monday, December 13, 2010 2:13 AM
>> To: Young, Milan
>> Cc: public-xg-htmlspeech@w3.org
>> Subject: Re: "Protocol" requirement - Interpretation
>>
>> Why does this have to be part of the speech API? Remote text
>> interpretation can be done without using any speech APIs. I've
>> implemented that myself a couple of times using just XmlHttpRequest.
>>
>> /Bjorn
>>
>> On Fri, Dec 10, 2010 at 7:57 PM, Young, Milan<Milan.Young@nuance.com>  wrote:
>>> Summary - Web applications must be able to request NL interpretation based
>>> only on text input (no audio sent).
>>>
>>>
>>>
>>> Description - Recognizers commonly separate lexical speech token processing
>>> from its semantic understanding.  This often requires multiple passes of
>>> semantic processing.
>>>
>>>
>>
>>
>>
>> --
>> Bjorn Bringert
>> Google UK Limited, Registered Office: Belgrave House, 76 Buckingham
>> Palace Road, London, SW1W 9TQ
>> Registered in England Number: 3977902
>>
>
>
>

Received on Thursday, 16 December 2010 16:47:09 UTC