Speech Tech Vendors See Opportunities in Closed Captioning
The Federal Communications Commission (FCC) is holding hearings on new government mandates regarding closed captioning of TV programming to maintain quality standards for accuracy, timing, program completeness, and placement of closed captions. The regulations have not yet gone into effect, but speech recognition companies are already working to make sure that their offerings are compliant once they do.
Nexidia entered the realm of closed captioning in 2012 with the release of Nexidia QC for Caption and Language Verification. QC analyzes live broadcasts and recorded media in real time, using quality control technology for closed captioning, language verification, and video descriptions.
Drew Lanham, senior vice president and general manger of media at Nexidia, explains that the solution uses several technologies to verify that language spoken within the audio is correct to ensure the quality. The QC offering can also realign time problems to make sure that the captions displayed are in sync with the spoken words.
Lanham maintains that the upcoming FCC mandates represent opportunities for Nexidia's closed captioning business. "The FCC and other regulatory agencies around the world are creating new policies to ensure that captions are correct, properly timed, and properly placed in all types of programming," he says. "With all of the devices that need to be supported, all of the content available, and all of the places in a workflow where captions can become damaged, manual QC processes are prohibitively inefficient and expensive. Each new law represents another compelling reason to switch to the automated QC of captions."
Nexidia updated its closed captioning product in March with the launch of QC 2.1, which includes enhanced job reporting. The product uses Nexidia's same core analytic technology and serves a market that "needs enhancements," says Ryan Pellet, the company's chief strategy officer.
Closed captioning, Pellet says, "is a natural expansion for us. There are billions of hours of audio that some of our large clients have on the contact center side. To me, this is the same. It's...about understanding and realigning a market that's underserved. This is tangential and not a departure from what we've been doing."
Closed captioning has come a long way since it first appeared on just one PBS show in the early 1970s. The technology entered the mainstream in the 1980s, and became a requirement by the FCC for most television content in 1996. Closed captioning must be available on all TV content, including television shows, commercials, films, live broadcasts, and video on demand.
Small companies, such as CPC, have developed solutions that integrate speech recognition with real-time captioning software. According to the company, its YouCaption software is based on the same speech engine as Nuance Communications' Dragon Naturally Speaking and has similar accuracy. However, the CPC speech engine is augmented for live captioning to minimize time lags.
In March, Italian speech-to-text provider PerVoice formed a new company with TVEyes, a U.S.-based global broadcast monitoring firm that provides online, real-time search and indexing for television and radio broadcasts. The new entity, TVEyes Language Technology, is working on enabling TV and radio broadcasting monitoring in any language.
"TVEyes Language Technology, leveraging PerVoice proprietary technology for speech recognition, will enable us to add new languages to our platform with unprecedented speed and efficiency," said David Ives, CEO and founder of TVEyes, in a statement. "It is applicable to every spoken language, including regional accents, and the economics can work for nearly any size market. In addition, if a country or region becomes of urgent interest to a particular client or set of clients, we can add real-time media monitoring for that country and language with rapid turnaround."
Closed captioning has been mainly used to help the hearing-impaired, but other audiences also use the technology, such as viewers learning English as a second language and other disabled viewers seeking to improve reading and comprehension skills.
Nexidia QC v2.3 features caption retiming and expanded support for international formats and languages.
The deal integrates PerVoice's audio-to-text technology with TVEyes' broadcast monitoring services.