Broadcast Beat Magazine 2018 NAB Show Edition | Page 97

flows . Timed Text Speech , a new file-based autotranscription service for subtitling and captioning workflows from Telestream is one of these solutions . Using speech-to-text technology , machine learning , and the speed of cloud-based processing , Timed Text Speech instantly creates a text transcript of a video and populates it directly into a captioning project with the correct timing so that each phrase is displayed at the right time . Timed Text Speech dramatically saves transcription time by allowing operators to review the instantaneous transcription results while the speech engine is processing the rest of the video program .
application of contextual , human-like understanding for improved accuracy and clarity of captions and subtitles is just the beginning of what machine learning can offer .
Such a solution saves content creators time and money when producing captions and subtitles for TV or internet content with short turnaround times , or for short-form content such as promos . It is also an efficient solution for captioning service companies looking for new tools to speed up their service , and a cost-effective and easy-to-use automated transcription option for corporations , government organizations and educational institutions that are captioning for the first time .
With Timed Text Speech , users upload media files directly to the services portal . Because they only pay by the transcription minute , there is no wasted expense for idle transcription resources . When combined with a closed captioning software such as Telestream MacCaption / CaptionMaker , Timed Text Speech becomes part of an all-in-one transcription and subtitling / captioning tool that allows users to easily transcribe , transcode , convert , troubleshoot , modify , and deploy caption data alongside video — regardless of the delivery platform . Conclusion
Even with help from automated systems and third-party specialists , live and post-production captioning has long been an imperfect art . New requirements and regulations for captioning of online video content have made the need for affordable and accurate captioning even more critical . Using the power of the cloud and machine learning , automated speech recognition systems are bringing new capabilities and enhanced productivity to captioning workflows . As schedules and budgets get even more compressed , the
Broadcast Beat Magazine • www . broadcastbeat . com • 97