Dynamic language model

• US 9,418,143 B2
• Filed: 01/26/2016
• Issued: 08/16/2016
• Est. Priority Date: 06/21/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Moreno Mengibar, Pedro J., Cohen, Michael H.
• Primary Examiner(s): Baker, Charlotte M
• Application Number:
  US15/006,392
• Publication Number:
  US 20160140218A1
• Time in Patent Office: 203 Days
• Field of Search: 704/235, 704/231
• US Class Current: 1/1
• CPC Class Codes: G06F 16/211 Schema design and management
• CPC Class Codes: G06F 16/29 Geographical information databases
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/3346 using probabilistic model
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/387 using geographical or spatial information, e.g. location
• CPC Class Codes: G10L 15/005 Language recognition
• CPC Class Codes: G10L 15/14 using statistical models, e.g. Hidden Markov Models [HMMs] G10L15/18 takes precedence
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams
• CPC Class Codes: G10L 15/24 Speech recognition using non-acoustical features
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/0633 using lexical or orthographic knowledge sources
• CPC Class Codes: G10L 2015/081 Search algorithms, e.g. Baum-Welch or Viterbi
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for speech recognition. One of the methods includes receiving a base language model for speech recognition including a first word sequence having a base probability value; receiving a voice search query associated with a query context; determining that a customized language model is to be used when the query context satisfies one or more criteria associated with the customized language model; obtaining the customized language model, the customized language model including the first word sequence having an adjusted probability value being the base probability value adjusted according to the query context; and converting the voice search query to a text search query based on one or more probabilities, each of the probabilities corresponding to a word sequence in a group of one or more word sequences, the group including the first word sequence having the adjusted probability value.

Claims

1. A non-transitory computer storage medium encoded with a computer program, the program comprising instructions that are operable, when executed by data processing apparatus, to cause the data processing apparatus to perform operations comprising:

receiving a voice search query from a user device;

determining a query context associated with the voice search query;

determining, based on the query context, to apply one or more language model rules;

dynamically adjusting one or more probabilities in a base language model according to the applied language model rules to generate a customized language model;

using the customized language model to generate a text search query from the voice search query; and

providing for display on the user device, one or more search results responsive to the text search query.

2. A method comprising:

receiving a voice search query from a user device;

determining a query context associated with the voice search query;

determining, based on the query context, to apply one or more language model rules;

dynamically adjusting one or more probabilities in a base language model according to the applied language model rules to generate a customized language model;

using the customized language model to generate a text search query from the voice search query; and

providing for display on the user device, one or more search results responsive to the text search query.

3. The method of claim 2, wherein each language model rules includes one or more adjustment factors for adjusting a probability value of a word sequence in the base language model.

4. The method of claim 3, wherein a magnitude of each adjustment factor depends on query frequency information.

5. The method of claim 3, wherein the query context includes a location of the user device and wherein the language model rules include a geographic language model rule having an adjustment factor for a particular word sequence located in a particular geographic region.

6. The method of claim 5, wherein applying the one or more language model rules includes determining that the location of the user device matches the particular word sequence of the geographic language model rule.

7. The method of claim 2, wherein the query context includes a social group of the user and wherein the language model rules include social group language model rule having an adjustment factor based on information submitted by users in the social group of the user.

8. The method of claim 7, wherein information includes one or more queries submitted by users in the social group of the user.

9. The method of claim 2, wherein the query context includes event information and the language model rules include event language model rules having an adjustment factor for adjusting a probability value of the base language model.

10. The method of claim 9, wherein the event information includes one or more of time of day and weather information.

11. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving a voice search query from a user device;

determining a query context associated with the voice search query;

determining, based on the query context, to apply one or more language model rules;

dynamically adjusting one or more probabilities in a base language model according to the applied language model rules to generate a customized language model;

using the customized language model to generate a text search query from the voice search query; and

providing for display on the user device, one or more search results responsive to the text search query.

12. The system of claim 11, wherein each language model rules includes one or more adjustment factors for adjusting a probability value of a word sequence in the base language model.

13. The system of claim 12, wherein a magnitude of each adjustment factor depends on query frequency information.

14. The system of claim 12, wherein the query context includes a location of the user device and wherein the language model rules include a geographic language model rule having an adjustment factor for a particular word sequence located in a particular geographic region.

15. The system of claim 14, wherein applying the one or more language model rules includes determining that the location of the user device matches the particular word sequence of the geographic language model rule.

16. The system of claim 11, wherein the query context includes a social group of the user and wherein the language model rules include social group language model rule having an adjustment factor based on information submitted by users in the social group of the user.

17. The system of claim 16, wherein information includes one or more queries submitted by users in the social group of the user.

18. The system of claim 11, wherein the query context includes event information and the language model rules include event language model rules having an adjustment factor for adjusting a probability value of the base language model.

19. The system of claim 18, wherein the event information includes one or more of time of day and weather information.