FCC Prohibits Text-to-Speech

On January 17, 2012, in Alerts & Warnings 101, CAP, Emergency Alert System, by with SRA International

This post is Part 1 of 5 in a series of reports on the contents of the FCC Fifth Report and Order released on January 10, 2012, which amends the Emergency Alert System (EAS) rules to accommodate Common Alerting Protocol (CAP) messages.

As a follow-on to our preliminary synopsis of the FCC Report and Order, this five-part series will take a more in-depth look at the new rules. Two of the highlights of the new CAP EAS rules might be the best and the worst decisions made in the ruling, both of which deal with how CAP alerts are converted for presentation as legacy EAS messages on broadcast stations and cable systems. First the good news…

Although the Commission was originally not in favor of allowing the use of the enhanced text displays that could be derived from CAP alerts because it felt the difference between CAP-derived text and that of text derived from legacy EAS messages might lead to public confusion, the Commission was swayed by public comments on this topic and thus changed its position. The FCC in the final ruling not only allows the enhanced CAP text, its use is required “to the extent that such text is supplied by the alert initiator”. CAP converters, which the FCC calls “intermediary devices”, must also include this ability to display the enhanced CAP text, but those devices have until June 30, 2015 to comply. As the FCC itself noted, the enhanced CAP text will make EAS messages more accessible to all Americans, especially Americans with disabilities. This is truly a commendable turnaround by the FCC, and it is to be acknowledged for taking the public comments to heart.

Unfortunately, there is also bad news in the FCC actions regarding CAP alerts. EAS Rule 11.56 (a) (2) starts out sounding promising saying that the FCC adopts the EAS-CAP Industry Group (ECIG) Implementation Guide, which was another good move by the Commission, but it goes on to say, “(except that any and all specifications set forth therein related to using text-to-speech technology… shall not be followed).” The FCC has thus prohibited the use of text-to-speech (TTS) for assembling the legacy EAS audio messages derived from CAP alerts when no CAP audio file is provided. The discussion for this decision is in Paragraph 38 of the Report and Order, citing concerns as to whether TTS is “sufficiently accurate” for EAS use and FCC feels that different TTS software could produce differing audio messages from the same EAS message. Although this decision is attributed in Footnote 121 to a comment from an EAS equipment manufacturer, one would think the FCC’s interpretation was not the intent of the original comment. It will be interesting to see if EAS equipment manufacturers, either individually or as a group through ECIG, file a Petition for Reconsideration on this issue. Such a Petition can be filed within 30 days of the Report and Order being published in the Federal Register. Broadcaster organizations and others may wish to chime in on this issue as well. It would seem TTS is already well proven through its use for alerting on NOAA Weather Radio, as well as the preliminary deployment of EAS CAP units already in the field. Let us hope this FCC decision is reconsidered.

To read the R&O use this link.

Check back to AWARE for future reports on the new FCC CAP EAS rules coming soon.


One Response to FCC Prohibits Text-to-Speech

  1. “The FCC has thus prohibited the use of text-to-speech (TTS) for assembling the legacy EAS audio messages derived from CAP alerts when no CAP audio file is provided.”

    On the surface, this may appear to be bad, but in the near term, this may be the most generous option.

    Many of those in my community have a hard time understanding the current version of text to speech. In other words, us old folks can’t hear what the computer is saying.

    There’s also the issue of geographical differences in words. For example, is “soda” and “pop” the same as “soda pop” or “Coke”. If one were to write “I’d like a Coke and fries”, the computer will read that hearer may need more information, ex. “We don’t serve Coke, is Royal Crown Cola OK?”

    Given the National Incident Management System guidance on “plain language”, this may be the best option at the moment.


Leave a Reply