[EAS] Test-To-Speech must be allowed - NOW!

[Clay Freinwald]

Sean - 

Let me try and help with this one.....

If you are going to go with TTS, as we have been doing here in Washington
State since Oct 2010....You have to have some goals.  One of the primary one
is based on the frightful situation with legacy EAS practice...That is, the
Video crawl (TV or Cable) is derived from the 'Header Codes' and the Aural
message is an analog voice recording that migrates from the source of the
message thru various systems and/or stations to the station or system that
sends the message on to the public.     This system, more often than not,
sends one message visually and a different one aurally. This is because the
Header Codes have no  way to be specific. 

For example - Supposed there is a Haz-Mat situation that calls for an
evacuation of a 6 square block area in Madison County.   The aural portion
of the message will provide the specific geographic area of the event, while
the Header Code derived text message will come out as Evacuate the ENTIRE
county.  This is because the Madison County EOC operator choses an EVI Event
Code.   Can you imagine how this is viewed by those that have been hounding
the FCC for years to resolve this type of shortcoming with the existing
system?

Now let's see how we deal with that using TTS -

The responsible party in Jefferson County, knowing there is TTS, writes the
public warning message on a computer (Located anywhere) and sends it to the
CAP server .   He knows that it will be read by a Text-Reader and carefully
crafts the message so that it will be read correctly by broadcaster or cable
systems.   He also knows that the exact text will be displayed as a crawl on
TV screens in his area.  (Uniform messaging should be our goal) 

Now suppose they do as you suggest - 

1- Write a Text Message (to drive the crawls) 
2 - Choose an Event Code 
3 - Voice the Aural Message.

A - We are back to where we were in the first place, the message must be
voiced.
B - The EOC worker will not be able to do this from any computer located
anywhere and must go to the EAS Equipment (endec) to record the message
C - Assuming the voice message is recorded as an MP3, how are we going to
get this distributed via CAP?
D - Do we abandon CAP and distribute the aural message via Legacy EAS?
E - Do we have these messages (Text and Voice) arrive a different times
using different distribution systems?
F - How can we expect the endecs at Stations and Systems to sort this out?
G - How about those states that use Daisy Chains and LP Stations.  Are they
going to relay the MP3 File?

I agree there is probably nothing prohibiting having video facilities get
their text info from CAP, however to suggest that a hybrid system be used is
a bit much to expect.

Remember....We are doing all we can to encourage those sources of public
warnings to use the EAS and one of the best ways of doing this is by keeping
it simple.  ( we have a lot of experience here in Washington State with very
rapid acceptance of TTS by Emergency Managers etc) 

Also remember - We are doing all we can to encourage Broadcasters and Cable
systems to use EAS (remember beyond EAN's and RMT's, it is still voluntary).
Broadcasters have also embraced TTS as they prefer the accuracy of the
message and have demonstrated they would rather hear TTS than a voice
message generated by an off-mic EOC operator who is scared and can read well
etc.

Sean, we've been there and done that.

Now what is the FCC's concern - Well they have hinted that it has something
to do with -

1 - Differences between different manufacturers TTS systems 
2 - Training Government Workers on how to write for TTS
3 - Difficulty of TTS in pronouncing certain local names.

A couple of suggestions have been made - 

1 - Have all the manufacturers of EAS equipment submit their equipment to
standardized testing to insure that their Text Readers operate uniformly.
I personally feel that this is a great suggestion.    The FCC should not be
reluctant to take this step.    It is certainly reasonable for purchasers of
EAS endecs to expect that equipment being offered would perform in a like
manner.

2- Training for those that write for Text Readers should/could be offered
and standardized.

Hope this answers you questions - I am confident that you too will come to
appreciate all the work that has been done to move EAS forward and that
includes Text to Speech.    A lot of people have put a lot of work into this
technique, it is our sincere hope that the FCC will see fit to re-consider
their position.

Clay Freinwald
Chair, Washington State SECC