ALEXANDRIA, Va., Feb. 11 -- United States Patent no. 12,548,553, issued on Feb. 10, was assigned to Amazon Technologies Inc. (Seattle).
"Techniques for machine learning based playback" was invented by Oliver Dayun Liu (Irvine, Calif.), Wenbin Ouyang (Redmond, Wash.), Sergio Elizondo Focil (Culver City, Calif.) and Wondong Jang (Beverly Hills, Calif.).
According to the abstract* released by the U.S. Patent & Trademark Office: "Techniques are described herein for providing variable length text during playback of a video. Audio corresponding to a portion of the video may be identified and segmented by a neural network. A number of captions of varying length can be identified for each segment using a second neural network. The user may be pre...