Real-time captioning of live video streams revolutionizes performance targets for platforms needing OPEX relief.
BURNABY, CANADA, April 10, 2024, NETINT Technologies today announced an industry-first automated subtitling feature powered by OpenAI’s Whisper automatic speech recognition model trained on 680,000 hours of multilingual data. Whisper is integrated into the NETINT Bitstreams Edge media processing application that runs on the NETINT Quadra Video Server Ampere Edition.
NETINT Unveils Industry-First Automated Subtitling Feature With OpenAI Whisper
In keeping with the cost reduction mandate video services must operate under, the 96 core Ampere® Altra® CPU enables accurate, real-time transcription for live broadcast and video streams with a power implementation that is unmatched.
Quadra Video Server Ampere Edition is $19,000 USD and when running OpenAI’s Whisper integrated with Bitstreams Edge will support up to 30 simultaneously transcoded live channels each with five HLS or DASH packaged profiles. Codecs supported include any combination of AV1, HEVC, and H.264 for encoding, with decoding support available with VP9, HEVC, and H.264. Resolutions scale to 8K and include 10-bit HDR and advanced rate-control mechanisms like capped CRF.
“Combining NETINT’s video processing acceleration with Ampere AI technology and high-performance, power efficient processing enables industry-first innovations for video and streaming services,” said Sean Varley, Ampere’s Chief Evangelist.
“Running the most advanced AI models on CPU headroom freed up by NETINT transcode offload in a single best-in-class platform opens up low cost, real-time video transcription at unprecedented scale for CDN operators worldwide.”
“Today marks a key milestone in making live video inclusive,” said Alex Liu, NETINT’s Co-founder and COO.
“Fusing OpenAI’s state-of-the-art Whisper technology with our Smart VPU solutions is enabling a new era of cost-effective video captioning with AI. Quadra Video Server with Whisper is years ahead in performance and computing efficiency which unlocks opportunities for broadcasters and content platforms to add subtitling without hesitation.”
OpenAI’s Whisper advanced speech-to-text models deliver industry-leading transcription precision, even in complex audio environments. Supporting multiple languages and a custom terminology lexicon, live streams are now more accessible to diverse global audiences with the Ampere Altra Processors.
In it’s first public showing, Ampere and NETINT are demonstrating Whisper running on the Quadra Video Server at the National Association of Broadcasters Show in Las Vegas, April 14th to April 17th in the NETINT booth located in the LVCC West Hall #4234. Interested companies are encouraged to request a meeting here.
Alternatively, there is a limited opportunity to join 50 industry experts for a private showing and Q/A session. For details and to register, click here.
About NETINT Technologies
NETINT Technologies is an innovator in ASIC-based video processing solutions delivering unparalleled transcoding performance, density and energy efficiency. Pioneering green computing approaches for video, NETINT is shaping the future of hyper-scale media encoding and processing with the first commercially available VPU (video processing unit). Learn more at netint.com.
About Ampere Computing
Ampere is a modern semiconductor company designing the future of cloud computing with the world’s first Cloud Native Processors. Built for the sustainable Cloud with the highest performance and best performance per watt, Ampere processors accelerate the delivery of all cloud computing applications. Ampere Cloud Native Processors provide industry-leading cloud performance, power efficiency, and scalability. For more information, visit Ampere Computing.
