Authenticity seal for video segments showing a human speaker
AssigneeEmovid Corporation
InventorsVictor Cho, Rupali Pathania, Samuel Nathan Brown
Life of this patent
Estimating the term…
Abstract
A facility for confirming the authenticity of an audio/video sequence showing a human speaker is described. The facility causes the audio/video sequence to be presented. In connection with presentation of the audio/video sequence, the facility presents a visual indication specifying an extent to which video and/or audio captured from the person was modified to produce the audio/video sequence.
Claims29
One or more computer memories collectively having contents configured to cause a computing system to perform a method, the method comprising:
- accessing an original audio/video sequence created by capturing (1) video images of a person speaking by an image sensor together with (2) audio of the speech by a microphone;
- for each of a plurality of forms of alteration of an audio/video sequence:
- accessing a control parameter specifying whether the form of alteration is to be applied to the original audio/video sequence;
- producing a resulting audio/video sequence from the original audio/video sequence by:
- applying the forms of alteration of the plurality of forms of alteration whose accessed control parameters specify that the form of alteration is to be applied to the original audio/video sequence;
- and
- adding a visual indication of authenticity, the visual indication of authenticity having a plurality of spatial zones, each of the spatial zones (1) corresponding to a different one of the plurality of forms of alteration, and (2) having a visual character indicative of whether the form of alteration was used in producing the resulting audio/video sequence.
A method in a computing system, the method comprising:
- causing an audio/video sequence of a person speaking to be presented;
- and
- in connection with presentation of the audio/video sequence, presenting a visual indication specifying an extent to which video and/or audio captured from the person was modified to produce the audio/video sequence.
15-20. (canceled)
One or more computer memories collectively having contents configured to cause a computing system to perform a method, the method comprising:
- causing an audio/video sequence of a person speaking to be presented;
- and
- in connection with presentation of the audio/video sequence, presenting a visual indication specifying an extent to which video and/or audio captured from the person was modified to produce the audio/video sequence.
A method in a computing system, the method comprising:
- accessing an original audio/video sequence created by capturing (1) video images of a person speaking by an image sensor together with (2) audio of the speech by a microphone;
- for each of a plurality of forms of alteration of an audio/video sequence:
- accessing a control parameter specifying whether the form of alteration is to be applied to the original audio/video sequence;
- producing a resulting audio/video sequence from the original audio/video sequence by:
- applying the forms of alteration of the plurality of forms of alteration whose accessed control parameters specify that the form of alteration is to be applied to the original audio/video sequence;
- and
- adding a visual indication of authenticity, the visual indication of authenticity having a plurality of spatial zones, each of the spatial zones (1) corresponding to a different one of the plurality of forms of alteration, and (2) having a visual character indicative of whether the form of alteration was used in producing the resulting audio/video sequence.
Dependencies are inferred from the claim text.