Accelerating the caption generation time and efficiency through Trance; the postproduction artificial intelligence (AI) captioning solution
The major governing body for professional and amateur tennis in the United States.
As a not-for-profit organization with more than 700,000 members, USTA invests 100% of its proceeds to promote and develop the growth of tennis from the grassroots to the professional levels. To that end, the association makes a broad array of high-value video content available for free streaming from its website and social media platforms. Content ranges from instructional videos and training courses to up-to-the-minute highlight reels from some of the most-watched events in the tennis world, such as the U.S. Open.
Content accessibility is a critical factor for any organization that serves the public, and the USTA is no exception. USTA is committed to making its content as easy as possible for all tennis enthusiasts to access, enjoy, and learn from. To that end, the association needed a fast, easy, cost-effective means of generating and embedding closed captions in video content.
The USTA’s previous captioning process was time-consuming and inefficient, relying on manual tasks and transcripts generated by a third-party speech-to-text service. Lacking an effective captioning integration tool for Adobe Premier Pro, video personnel were required to copy and paste text from the transcripts, line by line, onto the Premier Pro video timeline and manually enter timecodes.
Depending on the length of the video program, captioning was taking several hours – an unsatisfactory turnaround, especially for timely content such as the latest news from the U.S. Open. USTA wanted to speed up the captioning process and also free its technical personnel to focus on the creative aspects of their jobs.
To address these challenges, USTA has adopted Digital Nirvana Trance, a postproduction artificial intelligence (AI) captioning solution. Since moving to Digital Nirvana, the USTA has been able to cut the turnaround time for captioning an average video title from hours to only 30 minutes, with the captioning task completely offloaded from the technical team.
In a typical captioning workflow, a video technician simply drags the clip to a “hot folder” in the Digital Nirvana media service portal. The clip is then automatically uploaded to the processing center, where the Digital Nirvana team uses AI-based speech-to-text algorithms to generate a transcript, validate its accuracy, and then burn captions into the video. Once complete, the captioned video is automatically uploaded to USTA’s Premier Pro timeline.
Digital Nirvana’s Trance has met and exceeded USTA’s captioning requirements in several unique ways:
- A rapid turnaround time that is virtually unheard of in the captioning industry.
- Seamless integration with Adobe Premier Pro postproduction workflows.
- The industry’s highest caption quality standards, with captions that are 100-percent accurate and display with minimal delay as the text is spoken.
- Delivery of a complete product with captions burned into the video.
- Enterprise flexibility, with administrative features to support the most demanding requirements of any content provider.