US 11,900,700 B2 · Language agnostic drift correction

Abstract

Systems, methods, and computer-readable media are disclosed for language-agnostic subtitle drift detection and correction. A method may include determining subtitles and/or captions from media content (e.g., videos), the subtitles and/or captions corresponding to dialog in the media content. The subtitles may be broken up into segments which may be analyzed to determine a likelihood of drift (e.g., a likelihood that the subtitles are out of synchronization with the dialog in the media content) for each segment. For segments with a high likelihood of drift, the subtitles may be incrementally adjusted to determine an adjustment that eliminates and/or reduces the amount of drift, and the drift in the segment may be corrected based on the drift amount detected. A linear regression model and/or human blocks determined by human operators may be used to otherwise optimize drift correction.

Claims20

Dependent claims

A method comprising:

determining text data corresponding to dialog in media content, the text data including blocks of text data corresponding to the media content at time points;
analyzing at least a portion of the text data to determine a first likelihood of drift value;
determining the at least a portion of the text data includes drift based on the first likelihood of drift value, wherein drift is a misalignment between the at least a portion of the text data and the dialog at the time points;
adjusting, incrementally, a position of one or more of the blocks of text data in the at least a portion of the text data to determine adjusted text data;
determining a second likelihood of drift value for the adjusted text data, the second likelihood of drift value indicative of alignment between the adjusted text data and the dialog;
and
determining a correction amount to correct drift in the at least a portion of text data, the correction amount based on the adjusted text data.

4 more elements

A system comprising:

memory configured to store computer-executable instructions, and
at least one computer processor configured to access memory and execute the computer-executable instructions to:
determine text data corresponding to dialog in media content, the text data including blocks of text data corresponding to the media content at time points;
analyze at least a portion of the text data to determine a first likelihood of drift value;
determine the at least a portion of the text data includes drift based on the first likelihood of drift value, wherein drift is a misalignment between the at least a portion of the text data and the dialog at the time points;
adjust, incrementally, a position of one or more of the blocks of text data in the at least a portion of the text data to determine adjusted text data;
determine a second likelihood of drift value for the adjusted text data, the second likelihood of drift value indicative of alignment between the adjusted text data and the dialog;
and
determine a correction amount to correct drift in the at least a portion of text data, the correction amount based on the adjusted text data.

6 more elements

Dependencies are inferred from the claim text.

Language agnostic drift correction

Life of this patent

Language agnostic drift correction

Life of this patent

Abstract

Claims20