Back to results
US 8,805,689 B2Granted
Methods and apparatus to generate and use content-aware watermarks
AssigneeThe Nielsen Company (US), LLC
InventorsArun Ramaswamy, Robert A. Luff
Life of this patent
Estimating the term…
Abstract
Methods and apparatus to generate and use content-aware watermarks are disclosed herein. In a disclosed example method, media composition data is received and at least one word present in an audio track of the media composition data is selected. The word is then located in a watermark.
Claims29
A method, comprising:
- obtaining a plurality of words present in an audio track of a media composition, the plurality of words to be audible during playback of the audio track;
- selecting at least one of the words based on weight values for a plurality of terms respectively corresponding to the words, and the corresponding weight values of the at least one of the words being greater than a weight value of the ones of words which is not selected;
- and
- encoding the selected word in a watermark.
1 more element
An apparatus, comprising:
- a data interface to receive a media composition;
- a word selector to select a first one of a plurality of words present in an audio track of the media composition, the plurality of words to be audible during playback of the audio track, the word selector to select the first word based on respective weight values for a plurality of terms respectively corresponding to the words, and a first one of the weight values corresponding to the first word being greater than a second weight value of a non-selected one of the words;
- and
- a watermark encoder to encode the selected word in a watermark.
1 more element
A tangible machine accessible storage device or storage disc comprising instructions that, when executed, cause a machine to at least:
- obtain a plurality of words present in an audio track of a media composition, the plurality of words to be audible during playback of the audio track;
- select at least one of the words based on weight values for a plurality of terms respectively corresponding to the words, and the corresponding weight values of the at least one of the words being greater than a weight value of the ones of words which is not selected;
- and
- encode the selected word in a watermark.
1 more element
Dependencies are inferred from the claim text.