As TV broadcasters shift from wary tire-kicking to increasingly enthusiastic adoption of AI technology, new ways of doing things are taking hold at lightning speed in news production and myriad other aspects of station operations across the country.

At the same time, getting AI right no matter what it’s used for is top of mind everywhere. “We have to be really sharp about what we do with the efficiencies we find,” Sinclair Chief Innovation Officer Scott Ehrlich says. “It’s not just about the availability of functionality; it’s about how people actually use it. So it’s one step at a time.”

Pragmatic Recognition
Still, the overarching question now is, where will a growing stampede toward cutting costs in a competitive marketplace lead? As worries about the downsides—including the impact on jobs—give way to pragmatic recognition of the risks that come with being left behind, the likelihood grows that once an AI application lends a competitive edge to its users, other station groups will jump on board.

“AI represents a massive shift in culture and technology that we’ve only tapped the surface of,” Jeff Zellmer, executive vice president of digital operations at Fox Television Stations, says. Acknowledging that he, like everyone else, has “no idea where it’s going,” Zellmer voices the widely held perspective that “this is an area of growth we need to understand and be involved in.”

Preserving jobs remains a top priority for Fox Television and every other station group we talked to for this article, but the long view casts a different light on the issue. “Honestly, if we aren’t working on learning how to use AI every day, we won’t be ready to use it for something that’s absolutely critical for us,” Zellmer says. “Not taking advantage of that could have a big impact on jobs and our company’s fortunes over time.”

Station owners across the country say they’re taking cautious, responsible approaches to making AI an essential element in the operational toolbox, freeing people to do more things rather than replacing them. In growing numbers, they are implementing internal expertise and positions on the org chart aimed at vetting solutions and coming up with new ways to use the technology.

Gray Media’s approach typifies the current state of mind. “We’ve been very cautious with AI as a company, because viewer trust is our North Star,” Lee Zurik, senior vice president of news strategy and innovation at Gray Media, says. “We’ve been smart and methodical in how we’ve looked at AI, and now we’re ready to roll out some applications.”

Where things go from here and how fast largely depends on where the rubber meets the road as stations find reliable suppliers of whatever they’re looking for. AI “is an extremely exciting opportunity,” Zurik says, but the viability of products touting AI support “is all over the lot.” Talking with “dozens and dozens of vendors,” Zurik’s team is finding “some products are really good and can help us, but others still have work to do on accuracy and other issues,” he says.

Multilingual Audio
Not surprisingly at this early stage of exploration, there’s a wide gap among station groups as to how far caution governs the application of AI in operational workflows, newsgathering and on-air presentations. One station’s use case can be another’s bridge too far, especially when it comes to using AI to assist with news production.

Perhaps most dramatically, some broadcasters have been willing to break new ground by taking advantage of AI’s ability to play a role on air in live broadcasting. One category of vendor solutions drawing a lot of attention comprises real-time, video-
optimized audio translation platforms. These software systems can generate multilingual news, sports and other live commentary by replicating the voices and reshaping lip movements of presenters speaking in one language for simultaneous output in one or more alternative languages.

Among the most extensive uses of this application we’ve seen so far came in late February, when Sinclair announced it was using live AI translation technology supplied by Deeptune to test delivery of news in Spanish from English-speaking newscasters over live YouTube feeds from stations in Baltimore, San Antonio, West Palm Beach, Fla., and Las Vegas.

As described by Ehrlich, the move follows Sinclair’s successful use of generative AI last year when the company began using AI-enabled language translation with “Petko Unfiltered,” a Tennis Channel series, and in newscasts for use under hurricane conditions in the Gulf states.

Now, Sinclair is weighing whether to take the next step with the use of the translation technology in OTA broadcasts. “We have great confidence in the technology,” Ehrlich says. “It’s not like it can’t get better, but it’s reasonably mature at this stage. So you look at it and say, ‘Am I adding enough to the user experience to make this beneficial?’ ”

E.W. Scripps has been asking the same question, testing translations of NBC affiliate KRIS-TV newscasts in Corpus Christi, Texas, for delivery in Spanish over local Telemundo affiliate KAJA.

“Scripps is working to roll out near-time translation systems that would enable our news and sales teams to reach new audiences,” said Christina Hartman, vice president of emerging technology operations for Scripps. “By leveraging AI, it’s now possible to deliver content and advertising tailored for specific audiences that we would not have had the resources to produce even a few months ago.”

Public broadcasters, however, are not waiting. In March, XL8, an AI-driven real-time translation technology company, announced that it had successfully commercialized its real-time solution to translate English captions to Spanish. This breakthrough, achieved in collaboration with the Public Media Venture Group (PMVG), marks the first commercial deployment of AI-based real-time translation in U.S. public broadcasting, according to XL8.

The project was carried out in collaboration with DigiCAP, the Korea Radio Promotion Association (RAPA), and PBS station WCTE in Cookeville, Tenn., with XL8 serving as the core provider of real-time subtitle translation technology.

If at First You Don’t Succeed…
As Fox’s Zellmer notes, live language translation technology is getting a lot of attention across the broadcast industry, but many companies, including his, aren’t ready to take the plunge. “We did test it but didn’t see anything significant to justify moving forward,” he says. “Some people don’t like it, so we’re looking at whether there’s a way to give viewers a choice,” meaning the possibility of creating a “language feature in an app for people to choose rather than deciding for them.”

But Fox has no compunctions about trying things out, he says, noting that about a year and a half ago the company even went so far as to try using AI avatars in a couple of overnight newscasts going out on all platforms, which quickly proved to be a bad idea. “It shows we’re willing to experiment, but if it doesn’t help viewers with better experiences, it’s not right for us,” Zellmer says.

More generally, Fox is taking advantage of AI wherever it makes sense as a way to add efficiencies to workflows, such as vendor solutions that employ AI to expand content metadata and expedite searches across massive archives, which is “a huge area that can help us.” Fox is also using AI to help developers with coding in app-building.

Whether or not they endorse AI for an on-air A/V role, ever more station groups see the technology as a major asset strengthening TV journalism as a credible, vital alternative to the tsunami of “alternative facts” on the internet. The state of mind about what’s at stake for an industry highly dependent on local news reporting is well articulated by Michael Newman, whose title as director of transformation with Graham Media Group aptly describes the moment at hand.

“I came here because I believe in what journalism means to the communities we serve,” Newman says. “If we do AI the right way, it allows us to amplify our journalistic message and community voice and helps us to bring journalism into the modern world in ways we desperately need.”

MADISON, Wis.—The Wisconsin Educational Communications Board (ECB), the public media hub in Wisconsin, partners with the University of Wisconsin to deliver television and radio content statewide through six TV stations, six TV translators and 39 radio stations, reaching audiences through PBS Wisconsin, Wisconsin Public Radio and our most recent partner, Milwaukee PBS.

We manage TV master control operations with the primary goal of ensuring that content is delivered to all stations with uninterrupted service. Automation plays a key role in making that happen, and we rely on Crispin’s RapidPlayX 2000 and CORE master control automation systems with the webRPX playlist tool to keep us on the air.

Out With the Old
Our previous aging automation was no longer supported by its manufacturer, prompting our search for a new, reliable system. Crispin’s technology stood out from the field.

We started running the Crispin automation in shadowcasting mode alongside our existing software to let the system prove itself, and within a few weeks, we were confident about cutting over to Crispin exclusively. Maybe an hour after that cutover, I remember our executive director walking down to see how things were going. I said, “Great, we’re all going out to lunch!”

The new system has increased our broadcast reliability and reduced user errors significantly. We’re comfortable knowing that the system is running and just works.

The Crispin system’s scalability is a benefit as we add more coverage for viewers statewide. In 2023, Milwaukee PBS approached ECB about expanding our Crispin system to include managing their master-control needs. We started by running similar shadow-casting tests and even before we cut over, we got our first big test.

Wisconsin’s governor was delivering a speech to be broadcast statewide. Milwaukee PBS was planning to use their previous master control, but soon realized they were more comfortable with ECB’s service. Even though it was still early in the test phase, they asked if they could take our Crispin playout on-air for the speech, and the broadcast was seamless with no issues. That was the historic day we went on the air in Milwaukee for the first time.

Disaster Recovery
Like all stations, we are always preparing for a natural disaster or event that takes us off the air. With Crispin’s support, we’ve established an off-site disaster recovery (DR) location on the University of Wisconsin campus.

Our initial DR goals were to just survive and stay on air with whatever we could pass through, starting from scratch and building up. We soon decided we’d rather immediately continue broadcasting media previously scheduled for air from our DR facility.

Crispin developed a new application for its AssetBase peer synchronization technology to create a separate, fully synced off-site database that ensures we can continue media acquisition and playout of our full programming schedule in a seamless cutover. Other than sending data from A to B, our primary and DR Crispin systems operate separately and really don’t know about each other.

It doesn’t need to be a full-blown emergency for this DR system to serve our broadcasting needs—if our primary location is doing maintenance work or there’s any situation where we would otherwise need to shut down, we can easily switch over to our alternate site and maintain on-air continuity.

Most broadcast errors are user-related and Crispin’s intuitive software interfaces have significantly reduced those instances. If we ever do experience any issue—system or user-related—we know Crispin’s support will provide the peace of mind and reliability essential for public media.

More information is available at crispincorp.com.

Who knew that former FCC Commissioner Nathan Simington could be so quotable? But he let some doozies fly June 12 at the 2025 ATSC Meeting and NextGen Broadcast Conference in Washington, D.C., that one can only hope reflects the future direction of the reconstituted commission.

Looking down the road at the broadcast business, Simington said he sees datacasting as a means for broadcasters to get into the business of connectivity in a big way.

“You’ve got three legs of the stool right now: You’ve got connectivity, which is more like a toothpick than a leg; you’ve got ad sales; and you’ve got retrans,” he said. “With the pressure on others, the connectivity business is the only one that has a clear engineering advantage over any plausible entrant.”

Highlight, underscore and boldface that critical clause: “the connectivity business is the only one that has a clear engineering advantage over any plausible entrant.”

Whether that’s connectivity to back up GPS data from vulnerable satellites, connectivity to reach internet of things (IoT) devices simultaneously with a single transmission or connectivity to offload certain traffic from unicast wireless networks, broadcasters have a technical advantage upon which they must capitalize.

What makes the (now) former commissioner’s ATSC keynote even more powerful is that elsewhere he demonstrates a clear understanding of what connectivity and datacasting can mean to the health of the broadcast business.

“Long-term, I look at datacasting as enabling broadcasters to switch their primary source of revenue from connectivity and away from ad sales,” he said. And elsewhere in the speech, he said, “There’s really no option but to allow datacasting to flourish.”

All of this should be music to the ears of Conrad Clemson, the new CEO of EdgeBeam Wireless, the datacasting joint venture of four major broadcast groups. It should be the same for any broadcaster who recognizes the competition for ad spend only gets stiffer with each passing quarter or wonders how long the retrans gravy train can continue as cord-cutters chip away at MVPD subscriptions.

As this is being written, Olivia Trusty, a Republican, has been sworn in as FCC commissioner, joining FCC Chairman Brendan Carr, a Trump nominee confirmed in 2017 and named chair with the incoming administration, and Democrat Anna Gomez.

Even with Trusty now on board, the agency is still two commissioners short—and perhaps most importantly, one shy of a quorum—one can only hope that eventually it will be constituted of members who see things a bit more like Simington and a bit less like former Chair Jessica Rosenworcel.

Three short years ago, Rosenworcel told an NAB Show audience that ATSC 3.0 was “the right framework for right now” and that the “voluntary market-based” transition—with no new channel assignments for simulcasting as was done during the analog-to-digital transition—gives “broadcasters the opportunity to experiment with this technology [3.0], develop use cases and try to figure out what use cases work at scale.”

Even if one were to accept that ATSC 3.0 at that time was enabling broadcasters to experiment—something that would have surprised major TV manufacturers committed to NextGen TV—it’s clear that the use case testing phase is over, and it’s time to clear the way to let that toothpick grow into a titan.

Broadcast sound is crafted from a variety of microphones so that the final/composite sound field replicates what your ears really hear and the information your brain processes as the sound environment. It is amazing that your brain, along with some directional head-pointing, creates and mixes the soundtrack of life, everything around you, with all of its glorious sonic details. Plus, the brain remembers what you think it all sounds like and what you like.

A sound field is a composite of all the acoustic environment’s sounds with all the different intensities, moshed with direct and reflected sound waves that all blend and interfere with each other to create an “acoustic gumbo.” This is the simple reason why it is impossible to capture all the details of a soundscape with a single microphone.

Basic microphone physics begins with a single transducer/diaphragm capturing the entire sound field. In theory, a single omnidirectional microphone should be able to capture the sound field—like your ear—however, there are significant diminishing returns the further you move away from the microphone diaphragm. An analogy—a camera lens can focus in on an object for a “close-up” while a microphone cannot.

Separating the Noise
Typical sound environments are full of noise, which attenuates the clarity and fidelity of a sound source. Go into a venue and just listen—there is a base level of sound from HVAC systems and ice coolers, but mostly from the diffusion and decay of many complex sound waves. The problem is that your audio capture must be above this base layer of noise.

Early sports sound depended on just a few microphones that often blended a pleasant holistic representation of the sound field. Shotgun microphones inherently overreach their targeted sound and detect background clutter, like spectators. Often, this resulted in a good balance of sport and spectator sound from a single microphone. It was not until surround and immersive sound that the audio mix needed more defined spectator zones, which required the spectator atmosphere and ambiance to be captured separately from the event-specific sounds.

Mono transducers inherently have “full omnidirectional pickup” and are crafted for separation by physical characteristics like housing, porting and mounting. For example, a directional microphone is designed to reject directions and frequencies either physically by construction and design or electronically. Construction and porting creates cancellation of certain frequencies in the sound field that may not be desired, while electronics are also used to shape aspects of a sound field through cancellation, equalization and electronic manipulation.

Further, with fast computers, there has been success with multicapsule microphones that are capable of specific pattern isolation known as “beam forming.” The beams are electronically mapped and able to track the focused microphone beam pattern on a moving object such as an athlete or a ball.

Additionally, combinations of microphone capsules are used to create the illusion of stereo, surround and even immersive sound.

There is no doubt that the materials and technology of modern microphones are capable of a close-to-perfect representation of a sound field; however, capture still seems to be a mystery. A mediocre microphone can sound decent if it is properly mounted. Unfortunately, I am surprised that I still see shotgun microphones flat on camera lenses—check out baseball.

How to Aim a Shotgun
A shotgun microphone is designed and constructed to have focused-forward reach where the microphone is pointed; however, this requires full 360-degree free-field sound, not reflective sound from the camera lens. When the shotgun microphone is placed flat on the camera lens there is a “boundary” boost, but there is also frequency alteration.

Microphones benefit from accurate and proper axis and direction orientation. When stereo shotgun microphones were introduced, there was some carelessness with the left/right and even the up/down orientation of the microphone, so the sound didn’t match the picture. In a quiet setup, this may be difficult to detect, but it is obvious when the microphone is mounted on a handheld camera.

Television sound is about presenting an appropriate enhancement to the picture subject to interpretation, taste and the technology of the day. Most sound mixers work hard to get those singular microsonics that are unmistakably the event, but the great sound mixers have clean capture and a knack for that proper sonic blend.

Even though the entire broadcasting signal chain is digital, the beginning and the end are still analog. We live in an analog world and we ingest analog sound and visuals and convert frequencies to brain waves. So when the A1 sound mixer asks for some more microphone, humor them, because you know why.

LANCASTER, Pa.—At Take One Productions, we’ve been producing professional video content from Lancaster since 1984. What started as a corporate video company has grown into something bigger, especially in our work creating long-form commercial spots for the Pocono Mountains Visitors Bureau (PMVB), and in 2023, when we took over full broadcast operations for the Pocono Television Network (PTN).

PTN is a 24/7 television network available on streaming, over-the-air and cable. Designed to promote the Pocono Mountains region, PTN presents original segments like “Pocono Perspectives” and “Pocono Mountains Magazine,” local weather, live webcams and features on small businesses and regional events. Since its inception in 2017, the network has gained significant traction, quickly outgrowing the capabilities of its original distribution provider.

More Than Playout
To support this growth, we needed a platform built for demanding broadcast television operations. Our checklist included a playout system that managed ingest, media storage, graphics and master control, all while delivering scalability and rock-solid reliability. And because our studio is in Lancaster and PMVB is two hours north in Stroudsburg, it had to be cloud-based.

All these requirements led us to Imagine Communications’ Versio fully integrated on-prem and cloud playout system with the Versio Control configurable master control dashboard.

We reviewed a lot of playout systems and most felt bare-bones. But when we saw Versio in action, we immediately knew it was different—providing a fully integrated, cloud-capable playout system that could do everything we needed and more. We didn’t want a system we’d outgrow in a year; we wanted something we could build on, and Versio has proven to be exactly that.

Today, Versio is the engine behind PTN’s master control. We’re using the Versio Control dashboard to switch between live production, clip-based feedback and live weather from the Poconos area. In addition, the dashboard plays a critical role in our workflow. Before we go live, we check the interface to confirm that everything is properly synced and running as it should. That final verification step is essential—and Versio has become our go-to tool for ensuring broadcast continuity.

No Downtime
Not only has Versio been incredibly easy to operate, its reliability has been exceptional. In more than two years of continuous 24/7 broadcasting, we haven’t experienced a single minute of downtime due to a system failure—that’s virtually unheard of in this industry. On the rare occasions when we’ve had a question or issue, Imagine’s support has been outstanding. We open a ticket and within minutes, we’re on a live call with a team that resolves the issue right then and there.

With the Versio platform’s scalability and cloud capabilities, we’ve extended PTN’s reach far beyond what was possible just a few years ago. We’ve expanded from a handful of local cable systems to nearly a dozen across Pennsylvania. We stream live and on-demand content to major streaming platforms, and we’ve also added over-the-air broadcasting in six markets—with more on the way.

For a regional tourism network, that kind of reach is a major step forward. And as we look at future possibilities such as launching a second PTN channel, we know we have a platform that’s ready to grow with us.

More information is available at imaginecommunications.com.

We tend to think LED lights will last forever—they don’t, although 50,000 hours, plus or minus, is a very long time compared to legacy light sources. Assuming some driver component doesn’t suddenly fail, LED lights die more subtly; they slowly fade away. The challenge is knowing at what point they need to be replaced.

LEDs don’t burn out the way incandescents do. Instead, they tend to degrade gradually like fluorescents, but without the telltale tube-end blackening. At some point, following many hours of service, they will wear out and need to be replaced. That point is now being approached by the earliest adopters of LED lighting.

The lighting industry’s benchmark for the end of service life is defined as when LED output drops to 70% of its initial brightness. This is called the “L70” standard, where the “L” stands for lumens, and the 70 is the remaining percentage of its original brightness. Once output drops to that level, they’re declared “burned out.”

That brightness criteria only tells part of the story about how LEDs die, however. LEDs don’t just fade away; their color balance and fidelity can change over time.

Tuning Into Colors
Bicolor and full-color LED fixtures produce their intended shades of light by mixing several spectrum-tuned LEDs in combination. The mechanism for generating usable light from an LED is this: High-frequency (blue and UV) light stimulates the phosphors to emit a lower frequency of light. Think of it as changing largely invisible energy into colors we can see. The various phosphor “recipes” are “tuned” to produce different colors. Ah, science!

The resulting blend of light can remain correct for a very long time, but not forever. That’s because the phosphors slowly break down from heat and material fatigue. This is less of an issue with remote phosphor lights because their light-emitting layer is some distance from the LED chip’s most intense energy. The trade-off is that remote phosphor lights are limited to a single color temperature.

As phosphors wear unevenly, the color changes. Those “white” presets no longer work as intended, and color fidelity drifts. Some manufacturers (like ETC) already compensate for LED “droop” (or reduced output) caused by heat by adjusting individual LEDs in response to fixture temperatures in order to maintain proper color balance. A similar approach might one day address phosphor aging as well.

First-generation LED studio lights were fairly primitive compared to what’s available today.

Early LED fixtures strained to meet our industry’s high standards for color quality. A major hurdle is that LED lights are discontinuous spectrum sources, much like fluorescent. Sawtooth gaps in their spectrum profile can make flesh tones look splotchy and unnatural. More “full-spectrum” lights put their energy into higher color quality at the expense of brightness. In the early days of LED lights, the efficiency was too low to do a decent job of both. That problem’s been solved.

Too Many Metrics?
As color quality has become more important, so has our need to evaluate it. In fact, there are so many different metrics that it can be confusing. Of the many competing scales for color fidelity, including TLCI (Television Lighting Consistency Index), SSI (Spectral Similarity Index) and TM-30 (my current favorite), it’s CRI (Color Rending Index) that’s probably most familiar. We’ll use that as we discuss standards.

Every respectable LED studio light sold today has an excellent CRI score in the mid to upper 90s on a scale of 100. Excellence is now so common that you can find 90 CRI “bulbs” for your table lamp at the grocery store. Our studio lights should be at least that good. So, how do we ensure our lights measure up?

While an illuminance light meter can measure intensity, you’ll need something like the Sekonic C-800-U Spectrometer to know if the color fidelity is up to snuff.

Consider it a pass/fail test—anything less than 90 CRI is a “fail,” particularly for the important R9, R13 and R15 chips, which impact flesh tones.

Few LED lights from 10 or 15 years ago will clear this hurdle because the bar has been raised. LEDs are brighter, more efficient and have better color quality than ever. Improvements in LED design mirror the early days of computer chips: Each new generation outperforms the last.

“Close enough for TV,” like “state of the art,” is a moving target.

Better Resolution Needs Better Lighting
The urgency to upgrade your lights will be influenced by what’s considered “good enough.” Nothing’s entirely obsolete if it can still provide some service. After all, when fluorescent studio fixtures were first introduced, 80 CRI was considered “fine.” Today, what defines “close enough” when your audience is watching on 4K OLED monitors?

Can I just upgrade my old fixtures?

From an end-user perspective, it would be great if you could upgrade light engines as they fail. Unfortunately, the industry doesn’t work that way. With few exceptions, such as some Brightline, Kino Flo or Rosco fixtures, there’s often no way to repair, let alone upgrade, the light engine.

That inability to upgrade, or even repair, is a neglected issue in the broader lighting industry. Sadly, most fixtures are destined to be trashed when they reach the end of their useful service. Manufacturers should be encouraged to do better by our willingness to pay the price for fixtures that can be repaired or upgraded.

If your older fixtures are still working, perhaps they can fill secondary roles on your next studio set—if not the more critical task of lighting talent. Using “old school” color-
correction gels can help extend their usefulness. But remember that 10-year-old fixtures are unlikely to make it to 20 unscathed. Should your lights fail the test today, it’s time to plan and budget for their replacement tomorrow.

Educating your management team on the predictable life cycle of LED lights can help dispel the notion that they last forever. People less involved with lighting may not yet understand how much the technology has transformed. It’s no longer just replacing a burned-out lamp; it’s replacing the entire light fixture—at least for now. That may change over time, but the reality today is that LED lights don’t simply die—they fade away.