System and method for delivering personalized content based on deep learning of user emotions
Assignee10Tales, Inc.
InventorDAVID J. RUSSEK
Life of this patent
Estimating the term…
Abstract
Disclosed are a system, method and software to associate attributes with digital media assets using deep learning. Digital media contains specific assets, such as audio, that can be replaced with other assets. The system, method and software allow for personalizing digital media content based in part on neural network analysis to learn a user's emotional state.
Claims20
A method for delivering personalized content based on learned emotions of a user, comprising:
- capturing, by a server, a plurality of user interactions performed over time with a digital media stream, wherein each user interaction is associated with a respective time point;
- correlating, by the server, each user interaction to an emotional state;
- learning, by a learning algorithm communicatively coupled to the server, based on the correlations, the emotional state of the user at each respective time point;
- and
- serving, by the server, a new digital media stream based on the learned emotional states.
A system for delivering personalized content based on learned emotions of a user, comprising:
- a server configured to receive a plurality of user interactions performed over time with a digital media stream;
- a content server configured to store the digital media stream and transmit the digital media stream to the server;
- a correlation algorithm communicatively coupled to the server, the correlation algorithm configured to correlate each user interaction to an emotional state;
- and
- a learning module communicatively coupled to the server, the learning module configured to learn the emotional state of the user at each respective time point based on the correlations,
- wherein the server is configured to serve a new digital media stream based on the learned emotional states.
A method of delivering personalized content based on learned emotions of a user, comprising:
- capturing, by a server, a plurality of user interactions performed over time with a video stream, wherein each user interaction is associated with a respective time point;
- correlating, by the server each user interaction to a like or a dislike;
- learning, by a learning algorithm communicatively coupled to the server, based on the correlations, an emotional state of the user at each respective time point;
- and
- serving, by the server, a new digital media stream based on the learned emotional states.
Dependencies are inferred from the claim text.