Industry Insights: Teleprompter systems adapting to evolving newsroom workflows

As part of the ongoing Industry Insights series, we convened a roundtable discussion with leading vendors and suppliers in the broadcast industry to explore teleprompters.

This discussion covers key topics including the integration of SMPTE 2110 standards, the influence of artificial intelligence and machine learning on prompting systems, the rise of cloud-based solutions and the challenges of accommodating last-minute script changes during live broadcasts.

Participants also delve into the future of teleprompter technology and how it aligns with broader trends in remote production and newsroom integration.

Key takeaways from the Industry Insights roundtable

• SMPTE 2110 integration allows teleprompters to fit seamlessly into IP-based production workflows, reducing the need for external video converters.
• AI and machine learning are enhancing teleprompter capabilities, including voice recognition and automated scrolling, improving operational efficiency.
• Cloud-based prompting solutions offer significant advantages for remote and distributed production teams, simplifying deployment and connectivity.
• Modern prompting systems can quickly accommodate last-minute script changes during live broadcasts through real-time updates with newsroom computer systems (NRCS).
• Voice-activated prompting systems are providing more flexibility and ease of use for presenters, supporting ad-libbing and dynamic presentation styles.

How is SMPTE 2110 impacting your products?

Kamil Pietrzyk, support and projects manager, CueScript: SMPTE 2110, with its benefits of traffic shaping and precise delivery timing for uncompressed video streams over standard baseband, has gained significant popularity in the past two years and is set to become the go-to standard for all production teams. CueScript adopted ST2110 in 2019, incorporating it into both ends of their teleprompting workflow: the source CueB IP video rendering engine and the destination CSM range of prompter monitors. We quickly realized the importance of native and unified studio equipment workflows across all video layers, such as camera feeds and multiviewers, where teleprompting equipment should not be treated any differently nor require any unnecessary external video converters.

How are AI and machine learning being integrated into prompting systems?

Kamil Pietrzyk: While AI has minimal direct impact on teleprompting equipment or software, reporters and scriptwriters can significantly benefit from AI-assisted tools when preparing creative content for newscast stories. In contrast, machine learning is effectively used for the development, continuous improvement, and training of CueScript’s SayiT voice recognition engine. This engine listens to the anchor’s voice and automatically and intelligently controls the scroll speed of the teleprompting text.

Brian Kennedy, product manager, Ikan: AI tracking is a key feature in modern PTZ teleprompters that significantly enhances the production quality of presentations and broadcasts. This technology leverages artificial intelligence to monitor and respond to dynamic environments where multiple presenters may be speaking. AI systems in PTZ teleprompters detect the active speaker by analyzing the video feed to identify someone via facial recognition, body recognition, or whose lips are moving. This allows the camera to automatically focus on the person speaking, creating a seamless viewing experience for the audience.

Phil Dalgoutte, product director for prompting, Autoscript: Only with the advent of AI and machine learning based voice recognition has the possibility of true voice led teleprompting become possible. The prompting application fully automates these technologies, enabling the automatic scrolling of prompting text as the talent reads. It also allows voice control of the system for rundown navigation, thus automating the role of a prompt operator in many types of production.

How do cloud-based prompting solutions benefit remote or distributed production teams?

Kamil Pietrzyk: The concept of an easy remote teleprompting workflow has always existed, but the pandemic significantly accelerated the need for it as anchors suddenly had to broadcast from their living rooms. The primary challenge has been connecting studios/PCRs and remote location IP networks using traditional VPNs over public internet, which is challenging for broadcast IT teams and always raises security concerns. CueScript addressed this issue with its CueTalk Cloud solution, which radically simplifies teleprompter equipment deployment. By configuring the device, such as an IP monitor or CueVue, just once, it can be plugged into any public internet router and connected to the teleprompting software back in the PCR within seconds, allowing the operator to be in full control of the script and scroll speed ensuring metadata end to end encryption.

Advertisement

How do modern prompting systems accommodate last-minute script changes during live broadcasts?

Kamil Pietrzyk: In standalone prompting software, changes are managed by the operator within the Editor interface and these adjustments must seamlessly translate to the prompter output without delay. When integrated with NRCS systems via MOS protocol or other methods, the newsroom system continuously updates the prompting software in real-time, enabling the news team to make any necessary script changes, which reflects on the prompter monitors instantly. CueScript’s CueIt software further innovates with Widgets, providing anchors with customizable, real-time overlays for alerts and timers, all while maintaining script visibility, thus supporting dynamic and responsive news broadcasting.

Brian Kennedy: Scripts created or edited in the NRCS are automatically updated in the teleprompter software. Any last-minute changes made by the editorial team are immediately reflected, reducing the risk of discrepancies during live broadcasts. As soon as an editor makes changes to a script in the NRCS, those updates are reflected in the teleprompter software in real time. This allows news anchors to have the most current information, even if changes occur moments before going live.

Phil Dalgoutte: Last-minute changes are an intrinsic part of live production, particularly news. Typically, professionals manage this in two ways, and both are accommodated within Autoscript and Autocue prompting applications. Firstly, the prompting applications have a built-in script editor which can accommodate changes in any script that is currently open in the application. Secondly, within a news environment, the prompting application will directly connect with the newsroom control system, which handles the writing, compiling, and managing of all the scripts.

How do prompting systems integrate with newsroom computer systems (NRCS) for seamless workflow?

Phil Dalgoutte: Prompting systems connect to NRCS systems over an IP network connection. Alongside some bespoke implementations, several conform to the MOS (Media Object Server) scheme, alongside the prompting application itself. This connection allows the prompter to push rundowns and activate them. Alternatively, users can access rundowns from the NRCS within the prompting application by using a file tree style of navigation. The NRCS system maintains a live connection, automatically synchronizing any changes to the story ordering (rundown) or the story content to the prompting application.

How do voice-activated prompting systems compare to traditional methods in terms of accuracy and ease of use?

Kamil Pietrzyk: The primary challenge in voice-activated prompting systems has been enabling presenters to ad-lib or slightly deviate from the script and return to it effortlessly. CueScript’s second-generation SayIt, powered by machine learning, has this capability allowing presenters to have the confidence to improvise between script lines, enhancing their flexibility and spontaneity during broadcasts, knowing the prompter will begin scrolling again when they return to script. Operators and anchors traditionally rely on physical scroll controls and CueScript’s SayIt voice-activated engine is still complemented by these, offering a backup option to seamlessly switch between voice-activated control and traditional methods.

Phil Dalgoutte: Voice-activated systems have been in place for over ten years but could only activate and pace the scrolling of the script based on the pace of the talent speaking. There was no understanding of what was being said, making ad libbing impossible. True voice recognition-based prompting was first commercialized by Autoscript in 2020 and has been revolutionary. Now talent can drive a complete show with just their voice, handling ad libs, rephrasing, and free-form interview sections with ease.

What question about prompting should we be asking?

Kamil Pietrzyk: The main questions should cover topics like compatibility and integration with the existing and new studio equipment, remote connectivity options, the support and training available for the customer during and post purchase, and future developments acknowledging customer requirements. It’s essential to choose a company that not only provides quality products and expert service, but also listens to their customers. For any mission critical equipment, it’s vital to have a company you can rely on, which is why CueScript have a dedicated support team, available 24/7, specifically for the customer and their needs.

John Veen, Tekskil: The word prompter is a bit misleading — it’s really a matter of providing support to the on-air performer. A well supported person will deliver a dramatically better performance; the viewing audience is much less likely to switch the channel. And appreciative viewership is the bottom line for the business creating the show

How are virtualization and the cloud impacting your products?

Brian Kennedy: By using real-time ray tracing and other advanced technologies, we can accurately simulate various aspects of teleprompter performance. One critical area of focus is minimizing the amount of light entering the teleprompter hood. Excessive light can lead to unwanted reflections in the teleprompter glass, compromising the quality of the display. Simulations help us determine the best materials to use inside the hood.

Phil Dalgoutte: The way in which broadcast customers are looking to deploy, license, manage and operate products/software is changing significantly in the Tier 1 space (cloud, remote, flexible/dynamic licensing for scalability) so Autocue and Autoscript software architecture has evolved to meet those needs. It is now commonplace for prompting software to be hosted on-prem within a virtualized environment and even within the cloud for temporary production environments that need to spun-up just for particular broadcast events.