Southfield, MI, May 9, 2018 – The convergence of AV and IP brings a wealth of new opportunities to organizations as to how they produce and manage media, whether live or repurposed. ENCO’s latest advances to its patented captioning solutions, to be unveiled at InfoComm (June 6-8, Las Vegas Convention Center, Booth C4167), come to market at a time when organizations seek to incorporate live captions into their AV operations with greater frequency and efficiency – and easily find and repurpose existing content for online audiences.

At InfoComm, ENCO will unveil the now-shipping enCaption4 to the AV industry. With its powerful, real-time, automated speech-to-text engine, enCaption4 offers a robust open captioning solution for live presentations, seminars, lectures and sermons. enCaption4’s flexible architecture offers customers on-premise or cloud-based options to generate live open captions over top of video. This flexibility ensures that enCaption4 is applicable in both legacy and IP-enabled AV workflows, whether processing speech through a microphone for a single location; or distributing captions enterprise-wise from a cloud server over the IP network.

ENCO’s recent announcement of NDI® compatibility also makes enCaption4 well-suited to IP-based NDI video production and streaming environments. Commercial AV facilities with an NDI infrastructure can automatically generate captions with enCaption4 through its NDI input stream, and output an NDI signal with captions keyed directly on top of the video stream. This capability simplifies the captioning workflow by eliminating the need for specialized encoding hardware.

“There’s no question captioning workflows for live, pre-recorded and even long-form content have also been greatly simplified as more organizations transition AV operations to IP networks,” said Ken Frommert, president of ENCO. “enCaption4 is built to help these organizations take advantage of these new opportunities for hearing-impaired, multi-lingual and other audiences with minimal to no manual intervention, and at an exceptionally efficient price point.”

Repurpose with StreamGuys

enCaption4’s cloud option also presents new opportunities for captioning live and existing content for online audiences. In a new partnership with StreamGuys, a content delivery network and streaming media provider that will co-exhibit at ENCO’s booth, the two companies will show how businesses can extend the value of open captioning outside the local facility. Pairing StreamGuys’ SGrecast live recording and repurposing platform with enCaption4, users can generate automatic transcriptions of new recordings, and caption pre-recorded, archived files for streaming to mobile and the web. Users also have the power to export the captioned files (SRT or VTT), and save transcripts of any selected media content for later use.

“Once ingested into the system, SGrecast’s highly granular search capabilities will provide operators with unlimited flexibility to find captions that align with newly created video and audio files,” said Eduardo Martinez, director of technology, StreamGuys. “Imagine the flexibility this provides in government for court reporting or community meetings, where users can very quickly archive, search, find and recast content tailored to very specific needs and audiences. This will also prove highly valuable for corporations and universities with large volumes of archived media that can be used for training, education and other purposes.”

The integration also enables the ability to label and search for specific speakers. For example, enCaption4 provides generic speaker labels in their transcriptions. SGrecast users can label speakers for improved recognition and tracking, and later search the system for all content that contains a specific speaker, including exact spoken sentences.

Fast and Accurate

Frommert adds that feature sets in captioning solutions have expanded as speech-to-text conversion has grown faster and more reliable.

“enCaption4 integrates several breakthrough technologies that make it ideal for serving a broad variety of audiences,” said Frommert. “For one, our multi-speaker identification isolates and processes separate microphone feeds to reduce confusion from cross-talk in live meetings and seminars. enCaption4 also supports more than 20 languages, with extensive dictionaries that correlate with its groundbreaking accuracy.”

enCaption4’s high accuracy rate (more than 90 percent) is made possible through breakthroughs in machine learning technology, which have helped ENCO to develop a deep neural network approach to voice recognition. Frommert adds that the synchronization between speech and the on-screen appearance of captions has been reduced to about three seconds – a more than 50 percent reduction in latency from the previous generation of leading captioning solutions.