LAKE ORION, Mich.—For more than 20 years, Orion Neighborhood Television (ONTV) has been producing three public access channels serving the Lake Orion community, just north of Detroit. These PEG channels include a public access channel, education access channel, and government access channel carried by our local Comcast and AT&T U-verse systems. 

With diverse programming spanning local government meetings, school board meetings, sports, concerts, graduations and our own newscasts and public affairs programs, we produce roughly 800 shows a year. Using Cablecast Community Media software, we make this programming available via broadcast, livestream, and on demand, including our own YouTube channel with thousands of subscribers. Our channels are also accessible using Roku, Fire TV and Apple TV. 

As a 501(c)(3) non-profit entity, we are fortunate to have the generous support of our local government. We have a full HD production facility that we use to produce our shows, and for added revenue, we make it available for commercial use. We also teach production classes, offer internships to college students and co-ops to high school students, and provide grants to our school district’s video program to facilitate technology upgrades. 

 Committed to Captioning
We’re also committed to keeping our media facility up to date. For example, even though ONTV falls below the $3 million annual revenue mark—at which the FCC mandates captioning—we deem it important to make our programming more accessible to the deaf and hard of hearing. After evaluating our options, we chose the ENCO enCaption automated captioning solution. 

We have jumped years ahead in our captioning roadmap since installing enCaption, which we purchased through a county government grant. enCaption automates both live and non-live program captioning in real time with virtually no delay. It’s readily available whenever captions are needed, including for live coverage of government and school board meetings, among other live events. It is also far more affordable than using human captioners.

As meetings are produced and recorded, our Tricaster Mini at the venue feeds an SDI video signal into the Blackmagic 40x40 SDI router in our facility. That signal is then routed through our Cablecast system into enCaption. That program output then flows to our distribution channels, including Comcast, AT&T U-verse, and all our livestreaming platforms. Since our district doesn’t conduct meetings on Friday, we take that time to download our sidecar video files for the week and attach them to our VOD program files.

On-Site Captioning
For in-person attendees who are deaf or hard of hearing, we can also pump captions of the proceedings onto meeting room screens. This involves routing an NDI video signal from enCaption, with the captions over a black screen, back into the Tricaster on site. The auxiliary NDI output of the Tricaster is fed to a meeting control system, which displays it onto video screens next to and above the dais. If multiple people speak over each other, enCaption will zero in on the loudest and screen out others for clarity.

We also use ENCO’s DAD radio automation system to play out audio podcasts we produce, as well as PSAs, music and weather announcements onto our channels and Cablecast message boards. With up to 16 playback modules, DAD supports numerous workflows that include automated ingest, logging, scheduling, playout, programming control and more. 

Between enCaption and DAD, we’re very pleased with the high-quality results we get from these systems. With these reliable, affordable products, ENCO has become critically important to our overall production infrastructure. l

For more information about enCaption and/or DAD, visit www.enco.com.

NOVI, Mich.—ENCO has acquired  TranslateTV and Sentinel Solutions from Vox Frontera, a Silver Spring, Md.-based provider of televised Spanish language captioning and translation services. ENCO says the TTV acquisition will give the company a broader reach into the growing multigenerational Hispanic television market inside the U.S. and abroad by offering broadcasters an on-premises option for advanced translation of live Spanish-language TV captioning, with plans to develop a global solution specific to AV applications. Sentinel supports cost-effective, real-time quality monitoring solutions for automated captions—an especially valuable solution to keep broadcasters compliant with FCC closed-captioning standards related to accuracy, synchronization, and completeness, ENCO added. Financial details of the transaction were not disclosed.

TTV uses patented software to translate English closed captions into Spanish in real-time. While ENCO offers Spanish translation today within its flagship, automated enTranslate service—along with 45 other languages—enTranslate today only supports cloud-based processing for translation. Offering a fully on-premises captioning and translation system enables new business opportunities with customers that prefer to host their own systems and remain independent of the cloud. 

ENCO will leverage the same technology to develop an on-premises Spanish-language translation and captioning solution for financial institutions, corporations, government bodies and universities that restrict access to sensitive information. TTV’s open development platform will also allow ENCO engineers to make continuous speed and accuracy improvements, and add new languages moving forward for both broadcast and AV customers.

ENCO will natively integrate TTV into its enCaption automated captioning workflow in both broadcast and AV environments, which will allow the platform to be used for open captioning of meetings, classes, worship services and other AV-related events, according to ENCO President Ken Frommert.. 

“We look forward to supporting TTV’s legacy customers and bringing them into the ENCO family,” said Frommert. “Integrated with our enCaption automated captioning and transcription solution, we will enable this technology to natively accept live and prerecorded audio feeds and automatically convert them to text. Our same workflow will inject these translations onto multiple consumer platforms, presenting live, accurate captions on TV sets, desktop, laptops, and mobile devices. We will develop and innovate the TTV platform for new applications and use cases,  while bringing 24/7 product and technical support to the existing TTV client base.”

Frommert adds that with Sentinel, ENCO’s broadcast customers “can monitor and evaluate caption quality from ingest to transmission, while cleaning up inaccuracies, improving timeliness and providing documentation in response to FCC complaints.”

ENCO’s cloud-based enTranslate is available today as a plug-in to enCaption automated captioning and translation systems. Frommert expects to adapt a similar approach to TTV for its customers who want an on-premises version. All ENCO enCaption solutions leverage A.I. based deep learning and speech recognition algorithms “to deliver the most accurate and consistent live captions and translations in broadcast and AV,” the company said.

Gregory Schmidt, CEO of Vox Frontera, confirms that the company has transferred its TTV client base in the US, which includes The Tonight Show and local TV stations throughout the country, along with its TTV and Sentinel technologies. Longtime Vox Frontera Director of Engineering Dave Pinson will also join ENCO to support existing TTV customers and product development initiatives. 

“Vox Frontera has made a strategic decision to focus our company resources on other development efforts,” said Schmidt. “We know that ENCO offers best-in-class in automated broadcast captioning, and is a worthy successor that will serve our customers well.”

Above and beyond complying with FCC regulations, providing accurate and effective captioning for live content is becoming essential for broadcasters who want to stay relevant, competitive and ultimately successful in a changing media landscape. 

The provision of closed-captioned content has been an important part of many broadcaster requirements over recent years. However, while hearing impairment remains a leading basis for captions, with around 48 million hearing-impaired viewers in the United States, there are new drivers in the market. 

An increasing body of evidence suggests that audiences with unimpaired hearing enjoy watching content with captions. A quick online search generates dozens of articles commenting on this curious development—and expressing surprise that a growing number of people who don’t traditionally need captions to engage with content actually prefer captions with all their content. This is coupled with an increased need for captioning as more people watch content in louder environments or without audio, particularly those who are trying to keep up with a rapidly evolving news story. 

These trends are given further weight by research from social media platforms showing that captioning increases both content reach and engagement. With reach and engagement increasingly crucial to broadcasters, captioning content has become not just a mandate but a “no-brainer,” a value-add and a differentiator, especially in competitive local environments.

What makes things more complicated is an increasing demand for captions on live content. Traditionally, live captions are created by human operators to reach quality standards that allow major national broadcast networks, and their affiliates in the top 25 market areas, to live up to FCC guidelines. Outside those markets, broadcasters have been allowed to use the cheaper electronic newsroom technique (ENT), which converts a teleprompter script into live captions, but this method fails when it comes to breaking live events, sports commentary and other absolutely live uses.

The circle that needs to be squared for broadcasters to be able to reliably offer automatic live captions is making automatic caption generation as reliable as human captioning by stenographer. That means typically reaching around 98% accuracy within an acceptable latency between the spoken word and the matching caption appearing on screen, but at ENT costs.

OVERCOMING ISSUES WITH AUTOMATIC SPEECH RECOGNITION

AI-driven Automatic Speech Recognition (ASR) solutions are beginning to make headway in live automatic captioning, but still fail on many key metrics. They are hampered easily by background noise and multiple voices and, more importantly, they are yet to provide the combination of quality, economy and reliability that is really needed for live broadcasts—especially when it comes to sensitive topics such as politics, where every word counts. All broadcasters will be familiar with the potential consequences of an embarrassingly miscaptioned crucial word.

This is where ARC comes in. ARC stands for “Automatic Real-time Captioning” and is a new service-based method of optimizing ASR to bring its accuracy levels up to those of human performances. 

The best ASR engines are currently pushing around 95% accuracy and are managing to do so at roughly half the cost of human live captioning, bringing live captioning within reach of not just the top 25 markets, but broadcasters all over the country. With ARC, these engines are brought up to genuine human-level performance by adding a couple of powerful stages to the process, further refining the accuracy in the captioning.

The first of these stages is an optimization layer that uses a seed vocabulary manually curated by expert captioners, typically of names and places specific to a broadcaster’s individual market. The captioner expertise is key to achieving high levels of accuracy—there is a definite art to adding exactly the right vocabulary while avoiding false positives. This layer alone can reduce incorrect words in a live broadcast by up to 25% on average, taking automation from risky to consistent performance on the most important terms in a broadcaster’s output. 

The next stage of refinement comes when the ASR is run through a dedicated captioning platform. The platform adds all the features needed to put captions on screen properly and governs subtitle processing and presentation, as well as applying house rules and language formats for extra readability. This automated process makes use of existing data to guarantee consistent branding and channel identity in the captions, as well as preventing offensive language and other undesirable terms reaching the audience.

The result is an automatic captioning system that meets all accessibility requirements, with greater accuracy than any competing systems on the market and at a lower cost than traditional human live captioning. 

ARC is not tied to a specific ASR engine, allowing users to maintain their current technology investment or choose the best solution for their needs and piggybacking on the rapid developments in the field. It’s a flexible service solution; rather than paying upfront for a box that runs 24/7 regardless of content, ARC is available via a “pay-as-you-use” pricing model, significantly reducing per-station costs for TV networks.

FROM PAIN TO GAIN

ARC can turn automated live captioning in the U.S. from regulatory pain to commercial gain. It effectively makes the best better, taking the already impressive performance of the leading ASR vendors and making their captioning capabilities even more accurate more often, and at a cost that is far below the investment needed for traditional live captioning by stenographers or voicewriters.

Historically, captioning has been exclusively about providing services for hearing impaired people, but the broadcast landscape is changing. More people are watching more content on silent screens or in environments where sound is not always available or audible. Providing accurate, high-quality captioning is now becoming an important differentiator, especially when it comes to live news and sports content. Automatic Realtime Captioning with ASR technology is the way forward for broadcasters who want to stay relevant, competitive and ultimately successful in this context. 

Juliet Gauthier is strategic product manager, ARC, at Red Bee Media.

NEW YORK—The NAB Show New York will serve as the background for Enco to showcase its latest advances with machine learning for the enCaption4 automated live captioning system and enTranslate automated live translation systems designed to help hard-of-hearing and non-native-speaking audiences.

The enCaption4 is used to automate and augment live or pre-recorded content with closed or open captions in near real time. New features on display include advanced punctuation capabilities that enhance the accuracy and readability of transcriptions for full stops, commas, exclamation marks and question marks in English, German, Spanish and French. Enco describes the system as being able to detect the context of sentences to insert the appropriate punctuation and capitalization. Other new features offered are an enhanced scheduling interface, a web API for third-party integration and additional boosts to accuracy and speed.

The same speech-to-text engine in the enCaption4 is used in the enTranslate system, featuring advanced translation technology powered by Veritone. The enTranslate platform provides subtitles and secondary or tertiary closed captions in alternative languages, while users can also embed translated captions in VOD content or show live. It features a Neural Machine Translation methodology that blends AI and sophisticated linguistics modelling to provide translations based on context of current words and phrases. There are 46 languages supported by enTranslate.

Enco will display these two platforms at its booth, N355, during NAB Show New York, Oct. 16-17 at the Javits Convention Center.

Automation has been infused into innumerable elements of our daily lives. From production and assembly lines to broadcast facilities around the world, the transition to automated processes and workflows now have deep roots, and have forever changed the way we work, shop and entertain.

A common concern across all appliances of automation is the reduction, or outright elimination, of the human element. While the transition from manual to automated operations will undoubtedly remove human error in many cases, there are certainly more sensitive tasks where the argument for maintaining a manual workflow remains strong.

In the broadcast space, the transition to automated closed-captioning workflows is one topic that breeds intense discussion both for and against. However, the technology has advanced enough to instill confidence with broadcasters in many of today’s top DMAs, clearly representing the future of this important application.

EVOLUTION OF SPEECH-TO-TEXT

Speech recognition dates back to the 1950s, with modest first steps focused on digits and the most basic English words. With consumer services such as Siri and Alexa continuing to improve with each product generation, it’s clear that speed and accuracy in speech-to-text recognition has come a long way. So goes the same with automated captioning technology, which benefits today from the strengths of modern artificial intelligence.

[For an opposing view, read: Is It Live, Or Is It Automated Speech Recognition?]

While different mandates on closed-captioning in broadcast television exist around the world, the unifying purpose ensures that deaf and hearing-impaired viewers can fully understand and enjoy the shows they watch. Beyond the hearing impaired, statistics show that one in six viewers worldwide prefer to receive closed captions with their content.

Production and transmission of live, manual closed captioning has long been challenged by high costs, availability, varied latency, and inconsistent accuracy rates. And it’s true that the transition to more automated, software-defined captioning workflows introduced a new series of challenges.

For example, while automatic speech recognition removes the cost and staffing concerns of manual captioning, the performance of early-generation servers and processors demonstrated accuracy and latency problems. These issues were especially magnified for broadcasters that must now deliver accurate closed captioning across a multichannel, multilingual, multistandard and multiplatform media landscape.

These concerns are rapidly diminishing. The accuracy of speech-to-text conversion across multiple languages continues to improve with the emergence of powerful, deep neural network advances. In fact, accuracy across today’s strongest platforms has been raised to 90 percent or higher. The statistical algorithms associated with these advances, coupled with larger multilingual databases to mine, more effectively interpret—and accurately spell out—the speech coming through the air feed or mix-minus microphone.

Meanwhile, the faster and more powerful processing of computing engines within automated captioning technology has significantly reduced the latency to near real-time. This achievement is particularly impressive given that automated captions took between 30–60 seconds on many systems as recently as one or two generations ago.

Additionally, as closed-captioning software matures, emerging applications to eliminate crosstalk, improve speaker identification and ignore interruptions are improving the overall quality and experience for hearing impaired viewers.

MARCHING FORWARD

Many of the above improvements are related to recent breakthroughs in machine learning technology, which have enabled a deep neural network approach to voice recognition. Machine learning not only strengthens accuracy, it also provides value through detection of different languages and the different ways that people speak.

That intelligence as it relates to different dialects will provide an overall boost to accuracy in closed captioning. Consider a live news operation, where on-premise, automated captioning software now directly integrates with newsroom computer systems without the need for a network connection. This will now help broadcasters strengthen availability—no concerns about a network outage taking the system down—and take advantage of news scripts and rundowns to learn and validate the spelling of local names and terminology. Both of these points were once major and justified arguments against automated captioning.

Automated captioning also enables the applications to be achieved efficiently on a larger scale—costs are lowered due to the transition from human stenographers to computer automation. And as there is a need to caption a growing amount of content, there is an economy of scale that drives the cost down even further as broadcasters automate these processes.

As systems grow more reliable and broadcasters grow more comfortable with the technology, they will also find new efficiencies and opportunities along the way. For one, broadcasters that need to cut into a regularly scheduled program with breaking news or weather alerts will no longer be forced to find qualified (and expensive) live captioners on short notice.

Improvements in captioning technology have also been timely around emerging needs, including networks tasked with captioning large libraries of prerecorded content. As more systems move to software-defined platforms, the captioning workflow for prerecorded and/or long-form content has been greatly simplified. Post-production staff can essentially drag-and-drop video files into a file-based workflow that extracts the audio track for text conversion. These files can then be delivered in various lengths and formats for a TV broadcast, the web, mobile and other platforms.

And with multiplatform reach, broadcasters also have opportunities to caption live and on-demand streams, ensuring that hearing-impaired and multilingual audiences watching online are properly served as well. The future of this technology is very exciting, especially with the knowledge that we’re really just beginning to reap the fruits of this technology.

Ken Frommert is president of ENCO.

SOUTHFIELD, MICH.—United States patent number 7,047,191 B2 now belongs to Enco, per a company press release. Entitled “Method and System for Providing Automated Captioning for AV Signals,” the patent encompasses captioning technology and methods that help automate and accelerate speech-to-text translation to deliver open or closed captions; it specifically relates to Enco’s enCaption series of technology.

Enco’s software-defined enCaption systems provide accurate captioning for live and recorded content in near real time. The patent covers new advances that improve accuracy of enCaption’s speech-to-text engine, using advances in machine learning technology to develop a deep neural network approach to voice recognition.

This new technology includes a method that separates an audio signal from a combined AV signal and converts the audio signal to text data. The original AV signal is then encoded with the converted text data to produce, record and display a captioned AV signal. The patent also covers automatic translation of spoken words of a first language into a second language, with the translated words included in the captioning information.