[EAS] Best and Worst of New CAP EAS Rules

[Ed Czarnecki]

Bear in mind that the report and order opines that TTS conversion is
permissible at the source - i.e. at the EOC originator, as in the Perfect
Paul model.  The implied approach would be for a CAP message to contain a
resource link (URL) to an TTS audio file (mp3) that the CAP EAS device could
grab and download.

While permitting it at the source, the FCC order would prohibit TTS
conversion at the edge - at the CAP EAS device itself.  A cited concern
appears to be that of consistency from vendor to vendor.

I see this as inconsistent logic - the quality of TTS conversion could well
vary from originator to originator as well, depending on what system they
use (assuming the system they use even supports TTS conversion at the
source).   

Plus, TTS conversion at CAP EAS device level is at a minimum a useful (and
essential) backup in case the mp3 file is missing, corrupted, etc.,
regardless of whether the mp3 was recorded human voice or TTS conversion.

 

-----Original Message-----
From: eas-bounces at radiolists.net [mailto:eas-bounces at radiolists.net] On
Behalf Of Mike McCarthy

I read Harold Price's comment about bad speech conversion being less
preferred to no speech at all.

Upon a little reflection, I think that issue in part might have been an
underlying concern and they opted for now at least to stop the TTS train
from leaving the station.  Recall, the last major TTS deployment of that
scale was met the with frequent reference of "drunken Sweedish sailor".  
And nothing can be worse than conflicting voices or poorly pronounced
wordsin a critical situation.

I realize the vocoders of 2012 are vastly superior to the vocoder chosen
early last decade by the NWS. However, something got the FCC's attention
front and center for them to take such a drastic step when the train up to
that point was rolling strong with a spoken component as part of the plan.
And by this action, they're defining a breathing period to seek a
satisfactory resolution.

Once they're satisfied with whatever solution is defined and ratified by
consensus of the stakes-holders, they'll let that train roll again.

MM

On 1/18/2012 4:47 PM, Ed Czarnecki wrote:
> No, I think there is actually less to the story than some might think.
> Whether or not FEMA actually wanted voice converstion for ADA 
> compliance, several disabilities groups put information in the record 
> that argued against text to speech and speech to text.  Others have 
> been pressing the point at the Commission level on high.
>
> Also, I disagree with a suggestion that was made that the record 
> didn't reflect reliance on TTS by state systems.  CSRIC briefings, in 
> fact, did discuss that point (not the current CSRIC, but the prior 
> council I served on over a year ago).  Plus, FEMA had been providing 
> regular briefings to the FCC on its activities.
>
> Something got miscommunicated, to say the least.  But it can be fixed.
>
> Ed
>
> -----Original Message-----
> From: eas-bounces at radiolists.net [mailto:eas-bounces at radiolists.net] 
> On Behalf Of Mike McCarthy
>
> You know....with all the teeth gnashing about the text to voice 
> conversion prohibition, let me add one possible thought.

_______________________________________________
This is the EAS Forum Discussion List

Please invite your friends to join our Forum!
http://lists.radiolists.net/mailman/listinfo/eas

And, remember the main page: http://eas.radiolists.net