Bitmovin's Ai-Enabled Encoding Saves Time and Bandwidth
Bitmovin says they have been able to improve the speed and video quality from encoding operations through the use of artificial intelligence.
A standard encoding process involves performing an in-depth analysis of the entire video before encoding is started. Bitmovin’s AI-powered Encoding system works by continuously learning and recalling the parameters used in previous encodes, so that it can apply AI-optimised settings to each new video file.
Bitmovin's Encoding software is containerised, which makes it possible to scale the processing horizontally – the video file is split into chunks, and the chunks are then encoded simultaneously across as many different instances as desired. Containerisation gives users the control to scale up and down on the fly, as the workflow requires. By adding machine learning capabilities to the containerised software, Bitmovin is able to achieve faster processing times and higher output quality with no increase in bandwidth.
“Bandwidth should never hold back operators from delivering the best possible quality experiences,” said Stefan Lederer, CEO and co-founder at Bitmovin. “Artificial intelligence makes a notable difference in encoding, allowing operators to improve the visual quality of streams, eliminate buffering and improve consumer satisfaction.”
Bitmovin’s AI enabled approach allows each video to be handled more efficiently as well as delivering measurable quality improvements. The system's first pass is rapid high-level analysis, using machine learning to identify appropriate encoding settings as well as pre/post processing steps for each part of a video. The AI is trained using library content and will continue to learn over time.
On its second pass, the system makes a more detailed analysis of each part, fine tuning encoding parameters to ensure the best visual quality achievable for that file. Each section is analysed in parallel, cutting the overall time needed. The third pass applies what is learned from the analysis, using an optimized bitrate throughout the whole file, improving quality where possible and streaming at optimal bandwidth efficiency.
Bitmovin will continue to implement AI across their video delivery systems in the coming months. The company has pioneered in many significant developments in online video, including building and deploying one of the first commercial adaptive streaming (MPEG-DASH/HLS) HTML5 Players, to achieving some of the first 100x real-time encoding speeds in the cloud. Bitmovin customers include fuboTV, New York Times, Periscope, RTL, Sling TV and Bouygues Telecom - who collectively deliver several billion videos to consumers every day.All Bitmovin’s software is designed to be highly scalable, portable across any hardware and deliver high quality video. AI-enabled encoding will be demonstrated at the NAB Show 2018. www.bitmovin.com