System and method for assessing TV-related information over the internet
View Patent ↗The system retrieves information from the internet using multiple search engines that are simultaneously launched by the search engine commander. The commander is responsive to a speech-enabled system including a speech recognizer and natural language parser. The user speaks to the system in natural language requests, and the parser extracts the semantic content from the user's speech, based on a set of goal oriented grammars. The preferred system includes a fixed grammar and an updatable or downloaded grammar, allowing the system to be used without extensive training and yet capable of being customized for a particular user's purposes. Results obtained from the search engines are filtered based on information extracted from an electronic program guide and from prestored user profile data. The results may be displayed on screen or through synthesized speech.
1. A system for accessing supplemental network-resident information about an audio/video program comprising:
a network connection through which network-resident information may be obtained;
a speech recognizer receptive of a user's input speech request for information about a program;
a natural language parser coupled to said speech recognizer for extracting a semantic representation of the user's request for information;
a data store for storing a representation of an electronic program guide, said program guide having an indicator identifying the program being enjoyed by said user at the time of said user's request;
a search engine commander that issues at least one search request to at least one search engine through said network connection, wherein said search engine commander issues said at least one search request based on said semantic representation and using said representation of an electronic program guide as a first filter to constrain said search request to identify useful information relevant in determining what information the user is interested in retrieving;
a search results processor for receiving search results in response to said search request, for filtering said received search results based on the electronic program guide, and for providing a filtered portion of the received search results to the user as information about an audio/video program; and
at least one second filter, said second filter being used by said search results processor for selecting a first group of search results from said received search results and setting aside a second group of search results from said received search results, said second filter being customizable by the user; and
updating said information filters on a per-user basis based on historical data gathered as said use of the system.
2. The system of claim 1 further comprising, speech synthesizer coupled to said search results processor for providing the user with synthesized speech information about an audio/video program.
3. The system of claim 1 wherein said network connection provides connection to the internet.
4. The system of claim 1 wherein said search engine accesses at least one predefined site containing information about predefined topics pertaining to an audio/video program.
5. The system of claim 1 wherein said search engine commander includes a user profile data store for storing historical data about prior requests by the user for information.
6. The system of claim 1 wherein said search engine commander includes a mechanism for updating the contents of said electronic program guide data store.
7. The system of claim 1 wherein said natural language parser includes a set of predefined goal-oriented grammars.
8. The system of claim 1 wherein said natural language parser includes a data store for storing a set of grammars that are downloaded through said network connection.
9. A method of conducting a search for network-resident information about an audio/video program, comprising the steps of:
processing a user's input speech request for information about a program to generate a semantic representation;
formulating a search engine search request based on said semantic representation;
accessing a stored electronic program guide and using an electronic program guide to constrain said search request to identify useful information relevant in determining what information the user is interested in retrieving, said programming guide having an indicator identifying the program being enjoyed by said user at the time of said user's request;
sending said constrained search request to at least one search engine over a network;
receiving search results from said at least one search engine;
filtering said received search results based on the electronic program guide; and
filtering said received search results for selecting a first group of search results from the said received search results and discarding a second group of search results from said received search results, said second filter being customizable by the user; and
updating said information filters on a per-user basis based on historical data gathered as said use of the system.
10. The method of claim 9 further comprising providing said received search results to a speech synthesizer for generation of synthesized speech information.
11. The method of claim 9 further comprising periodically updating a data store containing said stored electronic program guide.
12. The method of claim 9 wherein said processing a user's input speech includes using a natural language parser.
13. The method of claim 12 further comprising downloading at least one grammar over said network and using said grammar with said natural language parser.