US 12,456,458 B2
Methods and systems for generating subtitles
Serhad Doken, Bryn Mawr, PA (US); Zhiyun Li, Kenmore, WA (US); and Dhananjay Lal, Englewood, CO (US)
Assigned to Adeia Guides Inc., San Jose, CA (US)
Filed by Adeia Guides, Inc., San Jose, CA (US)
Filed on Mar. 20, 2023, as Appl. No. 18/123,804.
Prior Publication US 2024/0321266 A1, Sep. 26, 2024
Int. Cl. G10L 15/18 (2013.01); G10L 15/06 (2013.01); G10L 15/08 (2006.01); G10L 15/22 (2006.01); H04N 21/488 (2011.01)
CPC G10L 15/18 (2013.01) [G10L 15/063 (2013.01); G10L 15/22 (2013.01); G10L 2015/088 (2013.01); H04N 21/4884 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method using control circuitry for generating subtitles comprising:
receiving utterance data corresponding to a media item;
generating first subtitles for the utterance data;
identifying a first term in the first subtitles;
determining contextual data relating to the utterance data, the contextual data comprising at least one of: user-generated comments, or historic subtitle data;
accessing a user profile comprising a plurality of user preferences;
determining, based on an automatic speech recognition (ASR) model trained with the contextual data, a replacement term for the first term, wherein the replacement term corresponds to at least one of the plurality of user preferences; and
generating second subtitles for the utterance data, the second subtitles comprising the replacement term.