Splitting term lists recognized from speech

• US 8,798,996 B2
• Filed: 03/05/2012
• Issued: 08/05/2014
• Est. Priority Date: 03/05/2012
• Status: Active Grant

• Current Assignee: Quotient Technology Inc.
• Original Assignee: Quotient Technology Inc.
• Inventors: Roloff, Jeffrey, Horowitz, Steven M.
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US13/412,479
• Publication Number:
  US 20130231918A1
• Time in Patent Office: 883 Days
• US Class Current: 704/257
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

In an embodiment, a method comprises analyzing a string of text that was generated based on audio input, identifying a plurality of text segments, wherein each text segment of the plurality of text segments comprises one or more words in the string of text, wherein at least one of the plurality of segments comprises a plurality of words, and organizing the plurality of text segments into a list of items, wherein each segment is a separate item in the list.

Claims

1. A method comprising:

determining a first context from among a plurality of contexts;

based on the first context, identifying a first technique for analyzing text data, wherein each context of the plurality of contexts is associated with a different technique for analyzing text data;

using the first technique to analyze a string of text that was generated based on audio data;

wherein using the first technique comprises identifying a plurality of text segments based on one or more criteria, wherein each segment of the plurality of text segments comprises one or more words in the string of text, wherein at least one text segment of the plurality of text segments comprises a plurality of words;

organizing the plurality of text segments into a list of items, wherein each text segment is a separate item in the list;

wherein the method is performed by one or more computing devices.

2. The method of claim 1, further comprising causing the list to be displayed on a computer display device.

3. The method of claim 1, further comprising, prior to analyzing the string, analyzing the audio data to generate the string of text.

4. The method of claim 1, wherein:

analyzing the string of text comprises identifying one or more pre-defined text within the string of text;

identifying the plurality of text segments is performed based on the location of the one or more pre-defined text within the string of text;

each text segment of the plurality of text segments is delineated based on the one or more pre-defined text.

5. The method of claim 1, further comprising:

based on one or more second criteria, identifying a plurality of locations within the audio input;

wherein identifying the plurality of text segments is based on the plurality of locations.

6. The method of claim 5, wherein identifying the plurality of locations is based on a length of a pause identified in the audio input.

7. The method of claim 1, wherein identifying the plurality of text segments comprises identifying the plurality of text segments based on a database of phrases, wherein each phrase in the database comprises one or more words.

8. The method of claim 1, wherein the first technique is associated with a first set of one or more database of phrases and a second technique that is associated with a second context of the plurality of contexts involves a second set of one or more databases that is different than the first set of one or more database.

9. The method of claim 1, wherein determining the first context comprises determining the first context based on (a) a client device that displays the list of items or (b) an application that receives the audio data from which the string of text is generated.

10. A method comprising:

analyzing a string of text that was generated based on audio data;

based on one or more criteria, identifying a plurality of text segments based on a database of phrases, wherein each phrase in the database of phrases comprises one or more words, wherein each segment of the plurality of text segments comprises one or more words in the string of text, wherein at least one text segment of the plurality of text segments comprises a plurality of words;

wherein identifying the plurality of text segments comprises;

identifying a first segment in the string of text;

determining whether the first segment matches a phrase in the database of phrases;

in response to determining that the first segment matches a particular phrase in the database of phrases;

identifying, in the string of text, a word that follows, in the string of text, the last word in the first segment,

adding the word to the first segment to generate a second segment, and

determining whether the second segment matches a phrase in the database of phrases;

wherein the method is performed by one or more computing devices.

11. The method of claim 10, wherein the first segment does not exactly match the particular phrase.

12. The method of claim 10, wherein:

the database of phrases is a first database of phrases;

the method further comprising, in response to determining that the second segment does not match any phrase in the first database of phrases, determining whether the first segment matches any phrase in a second database of phrases that is separate than the first database of phrases; and

each phrase in the second database of phrases comprises one or more words.

13. The method of claim 12, wherein:

the first database of phrases comprises an identification of a plurality of brand names; and

the second database of phrases comprises an identification of a plurality of generic items.

14. The method of claim 12, further comprising, in response to determining that the first segment does not match any phrase in the second database of phrases:

removing a word from the first segment to generate a third segment of one or more words;

determining whether the third segment matches a phrase in the second database of phrases.

15. The method of claim 12, wherein the first segment must exactly match a phrase in the second database of phrases in order for the first segment to match a phrase in the second database of phrases.

16. One or more storage media storing instructions which, when executed by one or more processors, cause:

determining a first context from among a plurality of contexts;

based on the first context, identifying a first technique for analyzing text data, wherein each context of the plurality of contexts is associated with a different technique for analyzing text data;

using the first technique to analyze a string of text that was generated based on audio data;

wherein using the first technique comprises identifying a plurality of text segments based on one or more criteria, wherein each segment of the plurality of text segments comprises one or more words in the string of text, wherein at least one text segment of the plurality of text segments comprises a plurality of words;

organizing the plurality of text segments into a list of items, wherein each text segment is a separate item in the list.

17. The one or more storage media of claim 16, wherein the instructions, when executed by the one or more processors, further cause causing the list to be displayed on a computer display device.

18. The one or more storage media of claim 16, wherein the instructions, when executed by the one or more processors, further cause, prior to analyzing the string, analyzing the audio data to generate the string of text.

19. The one or more storage media of claim 16, wherein:

analyzing the string of text comprises identifying one or more pre-defined text within the string of text;

identifying the plurality of text segments is performed based on the location of the one or more pre-defined text within the string of text;

each text segment of the plurality of text segments is delineated based on the one or more pre-defined text.

20. The one or more storage media of claim 16, wherein the instructions, when executed by one or more processors, further cause:

based on one or more criteria, identifying a plurality of locations within the audio input;

wherein identifying the plurality of text segments is based on the plurality of locations.

21. The one or more storage media of claim 20, wherein identifying the plurality of locations is based on a length of a pause identified in the audio input.

22. The one or more storage media of claim 16, wherein identifying the plurality of text segments comprises identifying the plurality of text segments based on a database of phrases, wherein each phrase in the database comprises one or more words.

23. The one or more storage media of claim 16, wherein the first technique is associated with a first set of one or more database of phrases and a second technique that is associated with a second context of the plurality of contexts involves a second set of one or more databases that is different than the first set of one or more database.

24. The one or more storage media of claim 16, wherein determining the first context comprises determining the first context based on (a) a client device that displays the list of items or (b) an application that receives the audio data from which the string of text is generated.

25. One or more storage media storing instructions which, when executed by one or more processors, cause:

analyzing a string of text that was generated based on audio data;

based on one or more criteria, identifying a plurality of text segments based on a database of phrases, wherein each phrase in the database of phrases comprises one or more words, wherein each segment of the plurality of text segments comprises one or more words in the string of text, wherein at least one text segment of the plurality of text segments comprises a plurality of words;

wherein identifying the plurality of text segments comprises;

identifying a first segment in the string of text;

determining whether the first segment matches a phrase in the database of phrases;

in response to determining that the first segment matches a particular phrase in the database of phrases;

identifying, in the string of text, a word that follows, in the string of text, the last word in the first segment,

adding the word to the first segment to generate a second segment, and

determining whether the second segment matches a phrase in the database of phrases.

26. The one or more storage media of claim 25, wherein the first segment does not exactly match the particular phrase.

27. The one or more storage media of claim 25, wherein:

the database of phrases is a first database of phrases;

the method further comprising, in response to determining that the second segment does not match any phrase in the first database of phrases, determining whether the first segment matches any phrase in a second database of phrases that is separate than the first database of phrases; and

each phrase in the second database of phrases comprises one or more words.

28. The one or more storage media of claim 27, wherein:

the first database of phrases comprises an identification of a plurality of brand names; and

the second database of phrases comprises an identification of a plurality of generic items.

29. The one or more storage media of claim 27, wherein the instructions, when executed by one or more processors, further cause, in response to determining that the first segment does not match any phrase in the second database of phrases:

removing a word from the first segment to generate a third segment of one or more words;

determining whether the third segment matches a phrase in the second database of phrases.

30. The one or more storage media of claim 27, wherein the first segment must exactly match a phrase in the second database of phrases in order for the first segment to match a phrase in the second database of phrases.