Self-learning methods, entity relations, remote control, and other features for real-time processing, storage, indexing, and delivery of segmented video
Self-learning systems process incoming data from sources such broadcast, cable, or IP-driven television and can discover topics that broadly describe the incoming data in real-time. These topics can be used to gather and store metadata from various metadata sources such as social networks. Using the metadata, content delivery systems working in parallel with the self-learning systems can deliver highly contextualized supplementary content to client applications, such as mobile devices used as “second screen” devices.
1 . A content delivery system for supplementing television content received from at least one television content source, the content delivery system comprising:
a capture platform operable to receive the television content from the at least one television content source and to process the received television content to generate at least one transcript corresponding to the received television content;
a topic extractor in communication with the capture platform, the topic extractor operable to receive the at least one transcript and to generate at least one topic by processing the received at least one transcript;
a metadata database operable to receive and store metadata from at least one metadata source providing metadata related to the received television content; and
a content enhancement engine operable to receive the at least one topic generated by the topic extractor and a portion of the metadata stored in the metadata database that corresponds with the received at least one topic, the content enhancement engine further operable to provide contextual content derived from the received portion of the metadata to at least one client device,
wherein the contextual content provided to the at least one client device supplements the television content received from the at least one television content source.
2 . The content delivery system of claim 1 , wherein the capture platform generates the at least one transcript by decoding audio data received from the at least one television content source using voice recognition.
3 . The content delivery system of claim 1 , wherein the capture platform generates the at least one transcript by processing a caption stream.
4 . The content deliver system of claim 1 , wherein the at least one client device is a mobile device that a user views simultaneously with the television content received from the at least one television content source.
5 . The content delivery system of claim 1 , wherein the contextual content comprises real time contextual content that is substantially aligned with real time television content.
6 . The content delivery system of claim 1 , wherein the contextual content comprises cached contextual content that is substantially aligned with television content previously received by the capture platform.
7 . The content delivery system of claim 1 , wherein the at least one television content source is selected from the group consisting of broadcast, cable, and IP-driven television.
8 . The content delivery system of claim 1 , wherein the content enhancement engine receives program scheduling information from an Electronic Programming Guide database.
9 . The content delivery system of claim 1 , wherein the metadata comprises at least one selected from the group consisting of substantially real time content feeds, news articles, and advertisements.
10 . The content delivery system of claim 1 , wherein the metadata database pulls the metadata from the at least one metadata source.
11 . The content delivery system of claim 1 , wherein the at least one metadata source pushes the metadata to the metadata database.
12 . The content delivery system of claim 1 , wherein the metadata database normalizes the metadata provided by the at least one metadata source, such that portions of metadata related to common topics may be merged or stored with a common identifier.
13 . The content delivery system of claim 1 further comprising an advertisement detector that detects adverts within the television content received from the television content source and provides advert data associated with the detected adverts to the content enhancement engine.
14 . The content delivery system of claim 13 , wherein the contextual content comprises at least one selected from the group consisting of the topics, the substantially real time content feeds, advert indicators, brands, people, places, programs, organizations, and stocks.
15 . The content delivery system of claim 1 , wherein the at least one client device accesses the contextual content through an application programming interface (API).
16 . An advertisement recognition system for detecting adverts within television content received from at least one television content source, the advertisement recognition system comprising:
a capture platform that captures the television content from the at least one television content source and extracts individual sentences from the captured television content;
an advert database;
an advert identification system in connection with the capture platform, the advert identification system operable to analyze the individual sentences in conjunction with the advert database to identify potential adverts and confirmed adverts; and
an advert validation system operable to receive the potential adverts from the advert identification system and further operable to confirm the received potential adverts, converting them to the confirmed adverts.
17 . The advertisement recognition system of claim 16 , wherein the at least one television content source is selected from a group consisting of broadcast, cable, and IP-driven television.
18 . The advertisement recognition system of claim 16 , wherein the advert validation system comprises a human administrator who manually confirms the received potential adverts as the confirmed adverts.
19 . The advertisement recognition system of claim 16 further comprising an advert filtration system that filters the confirmed adverts from search results provided to at least one client device.
20 . The advertisement recognition system of claim 16 , wherein the advert database stores the individual sentences with associated data.
21 . The advertisement recognition system of claim 20 , wherein the associated data includes count information representing the number of detected occurrences of the individual sentences.
22 . The advertisement recognition system of claim 20 , wherein the associated data includes linking information representing other individual sentences most frequently occurring immediately before or after the individual sentences.
23 . The advertisement recognition system of claim 16 , wherein the advert identification system identifies the potential adverts and the confirmed adverts as clusters of the individual sentences.
24 . The advertisement recognition system of claim 23 , wherein the advert identification system utilizes Markov models to identify the potential adverts and the confirmed adverts.