Background contextual conversational search

• US 8,037,070 B2
• Filed: 06/25/2008
• Issued: 10/11/2011
• Est. Priority Date: 06/25/2008
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Maghoul, Farzin
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): NGUYEN, MERILYN P
• Application Number:
  US12/146,227
• Publication Number:
  US 20090327263A1
• Time in Patent Office: 1,203 Days
• Field of Search: 707/736, 707/750, 707/755, 707/738, 707/748, 707/749, 707/758, 707/769, 707/767, 707/770
• US Class Current: 707/736
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A method of generating search queries based on digitized audio from conversations, including: providing a database having a global hot-list of universal popular keywords or phrases and a personalized entity list comprising keywords and phrases used with a frequency above a determined threshold value in conversations involving a user; monitoring a conversation between at least two people, including the user; identifying words or phrases in digitized audio of the monitored conversation through speech recognition; comparing the identified words or phrases to the keywords and phrases in the database to find any matches; generating a search string, without the user requesting a search, based on words or phrases found to match the keyword or phrases stored in the database; submitting the search string to a search engine as a search query; and serving a set of search results returned by the search engine to a display device of the user.

Claims

1. A computer-implemented method of generating search queries based on digitized audio from conversations, the method executable by a computer including a processor and memory, comprising:

providing a database stored in the memory containing a global hot-list comprising universal popular keywords or keyword phrases and containing a personalized entity list comprising keywords and keyword phrases used with a frequency above a determined threshold value in conversations involving a user;

monitoring a conversation between at least two people, at least one of which is the user;

identifying, by the processor, words or phrases in digitized audio of the monitored conversation through speech recognition;

comparing, by the processor, the identified words or phrases to the keywords and keyword phrases in the database to find any matches;

generating, by the processor, a search string, without the user requesting a search, based on words or phrases found to match the keyword or keyword phrases stored in the database;

submitting, by the computer, the search string to a search engine as a search query; and

serving, by the computer, a set of search results returned by the search engine based on the submitted search string to a display device of the user.

2. The method of claim 1, wherein the conversations comprise conversations over telephones, the method further comprising:

sensing a contextual piece of information within the monitored conversation or as related to a status of one or more of the telephones; and

integrating the contextual piece of information with the search string to narrow the submitted search query to the search engine, to thereby affect the search results thereof.

3. The method of claim 2, wherein the contextual piece of information comprises a location detected by a global positioning system (GPS) device located within the telephone, words uttered in proximity to the identified words within the search string, a time stamp, or a combination thereof.

4. The method of claim 1, wherein the conversations comprise conversations over telephones, the method further comprising:

storing in a search diary of the telephone of the user all of the results generated during the conversation; and

making available the search diary to the user through a display of the telephone.

5. The method of claim 1, wherein the conversations comprise conversations over telephones, the method further comprising:

digitizing an audio stream of a video or other media being played on the telephone of the user; and

identifying words or phrases in the digitized audio stream through speech recognition.

6. The method of claim 1, wherein the personalized entity list comprises words selected from the group consisting of movies, contacts in an address book, favorite restaurants, favorite news types, favorite travel destinations, favorite foods, celebrities, and combinations thereof.

7. The method of claim 1, further comprising:

tracking a number of times a word or phrase is used in conversations involving the user; and

adding the word or phrase to the personalized entity list once the number of times it is used rises above the determined threshold value.

8. The method of claim 2, wherein the contextual piece of information comprises words uttered in proximity to the identified words within the search string.

9. A computer-implemented method of generating search queries based on a media audio stream played over a media device of a user, the method executable by a computer including a processor and memory, comprising:

digitizing any analog words detected in an audio stream of a video or other media played over a media device of a user;

monitoring digitized conversations of the user and at least one other person;

providing a database stored in the memory containing a global hot-list comprising universal popular keywords or phrases and containing a personalized entity list comprising keywords and keyword phrases used with a frequency above a determined threshold value in the monitored conversations involving the user;

identifying, by the processor, words or phrases in the digitized audio stream through speech recognition;

comparing, by the processor, the identified words or phrases to the keywords and keyword phrases in the database to find any matches;

generating, by the processor, a search string, without the user requesting a search, based on words or phrases found to match the keyword or keyword phrases stored in the database;

submitting, by the computer, the search string to a search engine as a search query; and

serving, by the computer, a set of search results returned by the search engine based on the submitted search string to a display of the media device of the user.

10. The method of claim 9, wherein the conversations comprise conversations over telephones, the method further comprising:

sensing a contextual piece of information within the audio stream or as related to a status of one or more of the telephones; and

inserting the contextual piece of information with the search string to narrow the automatically-submitted search query to the search engine, to thereby affect the search results thereof.

11. The method of claim 10, wherein the status of the telephone comprises a location detected by a global positioning system (GPS) device located within the telephone, a time stamp, or a combination thereof.

12. The method of claim 9, wherein the conversations comprise conversations over telephones, wherein the telephone of the user includes the media device, the method further comprising:

storing in a search diary of the telephone of the user all of the results generated during playing of the video or other media; and

making available the search diary to the user through a display of the telephone.

13. The method of claim 9, wherein the personalized entity list comprises words selected from the group consisting of movies, contacts in an address book, favorite restaurants, favorite news types, favorite travel destinations, favorite foods, celebrities, and combinations thereof.

14. The method of claim 9, further comprising:

tracking a number of times a word or phrase is used in conversations involving the user; and

adding the word or phrase to the personalized entity list once the number of times it is used rises above the determined threshold value.

15. A system for generating search queries based on digitized audio from a conversation between at least two people, at least one of which is a user, the system comprising:

a memory to store computer programmable code;

a processor to execute the computer programmable code stored in the memory;

a database to store entity lists comprising a global hot-list with universally popular keywords and keyword phrases and an entity list with keywords and keyword phrases used with a frequency above a determined threshold value in conversations involving the user;

an identifier to identify words or phrases in the digitized audio of the conversation between the user and at least one other person through speech recognition;

a comparator to compare the identified words or phrases to the keywords and keyword phrases stored in the database to locate any matches;

a search string generator to generate a search string, without the user requesting a search, based on words or phrases found to match the keyword or keyword phrases stored in the database;

wherein the processor;

submits the search string to a search engine as a search query; and

sends a set of search results returned by the search engine based on the submitted search string to a display device of the user.

16. The system of claim 15, wherein the conversations comprise conversations over telephones, wherein the search string generator detects a contextual piece of information within the monitored conversation or as related to a status of one or more of the telephones;

and integrates the contextual piece of information with the search string to narrow the submitted search query to the search engine, to thereby affect the search results thereof.

17. The system of claim 16, wherein the contextual piece of information comprises a location detected by a global position system (GPS), words uttered in proximity to the identified words within the search string, a time stamp, or a combination thereof.

18. The system of claim 15, wherein the conversations comprise conversations over telephones, the system further comprising:

a search diary database to store a search diary of cumulative search results for the user over a period of telephone conversations, wherein the search diary is made available to the user through a website posted to the Internet.

19. The system of claim 15, wherein the personalized entity list comprises words selected from the group consisting of movies, contacts in an address book, favorite restaurants, favorite news types, favorite travel destinations, favorite foods, celebrities, and combinations thereof.

20. The system of claim 15, wherein the processor:

tracks a number of times a word or phrase is used in conversations involving the user;

increments a counter each time the word or phrase is detected; and

stores the word or phrase in the database once the counter for that word or phrase rises above a determined threshold value.

21. A system for generating search queries based on a media audio stream played over a telephone of a user, the system comprising:

a memory to store computer programmable code;

a processor to execute the computer programmable code stored in the memory;

a database to store entity lists comprising a global hot-list with universally popular keywords and keyword phrases and an entity list with keywords and keyword phrases used with a frequency above a determined threshold value in conversations involving the user;

a digitizer to digitize any analog words detected in an audio stream of a video or other media played over a telephone of the user;

an identifier to identify words or phrases in the digitized audio stream through speech recognition;

a comparator to compare the identified words or phrases to the keywords and keyword phrases stored in the database to locate any matches;

a search string generator to generate a search string, without the user requesting a search, based on words or phrases found to match the keyword or keyword phrases stored in the database;

wherein the processor;

submits the search string to a search engine as a search query; and

sends a set of search results returned by the search engine based on the submitted search string to the telephone of the user.

22. The system of claim 21, wherein the search string generator detects a contextual piece of information within the audio stream or as related to a status of the telephone;

and integrates the contextual piece of information with the search string to narrow the automatically-submitted search query to the search engine, to thereby affect the search results thereof.

23. The system of claim 22, wherein the status of the telephone comprises a location detected by a global positioning system (GPS) device located within the telephone, a time stamp, or a combination thereof.

24. The system of claim 21, further comprising:

a search diary database to store a search diary of cumulative search results for the user over a period of telephone conversations, wherein the search diary is made available to the user through a website posted to the Internet.

25. The system of claim 21, wherein the personalized entity list comprises words selected from the group consisting of movies, contacts in an address book, favorite restaurants, favorite news types, favorite travel destinations, favorite foods, celebrities, and combinations thereof.