System and method for inputting text into electronic devices

• US 9,424,246 B2
• Filed: 06/17/2014
• Issued: 08/23/2016
• Est. Priority Date: 03/30/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Touchtype Limited (Microsoft Corporation)
• Inventors: Spencer, Stephen Thomas, Creed, Pid, Medlock, Benjamin William, Orr, Douglas Alexander Harper
• Primary Examiner(s): He, Jialong
• Application Number:
  US14/307,308
• Publication Number:
  US 20140297267A1
• Time in Patent Office: 798 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 3/04886 by partitioning the display area of the touch-screen or the surface of the digitising tablet into independently controllable areas, e.g. virtual keyboards or menus
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models

Abstract

Systems comprising a user interface configured to receive text input by a user and a text prediction engine configured to receive the input text and generate text predictions. The text prediction engine may comprise a general language model and a context-specific language model. The text prediction engine is configured to generate text predictions from the general language model and the context-specific language model and combine the text predictions. The text prediction engine may comprise first and second language models and a first context-specific weighting factor associated with the first language model. The text prediction engine is configured to generate text predictions using the first and second language models, generate weighted probabilities of the text predictions from the first language model using the first context-specific weighting factor; and generate final text predictions from the weighted predictions generated from the first language model and the predictions generated by the second language model.

Claims

1. A system comprising:

a processor; and

memory storing instructions that, when executed by the processor, configure the processor to;

receive text input by a user;

generate at least one first text prediction, each first text prediction comprising a first term and a first associated probability value, based on the received text input, using a general language model;

generate at least one second text prediction, each second text prediction comprising a second term and a second associated probability value, based on the received text input, using a context-specific language model, wherein the context-specific language model is trained on user input text inputted during a specific time period;

combine the at least one first text prediction generated using the general language model and the at least one second text prediction generated using the context-specific language model; and

output one or more of the combined text predictions.

2. The system of claim 1, wherein the general language model is a dynamic language model.

3. The system of claim 1, wherein the context-specific language model is a dynamic language model.

4. The system of claim 1, wherein the context-specific language model further comprises a MessageThread-specific language model, a recipient-specific language model, or a location-specific language model.

5. The system of claim 1, wherein the specific period of time comprises at least one of:

a day, month, year, or weekend.

6. The system of claim 1, wherein at least one of the first text prediction or the second text prediction comprises a word candidate, and wherein when a word candidate generated using the general language model matches a word candidate generated using the context-specific language model, the processor is configured to apply a weighting factor to the probability of the word candidate generated using the general language model.

7. The system of claim 6, wherein the weighting factor is based on the probability of that word generated using the context-specific language model.

8. The system of claim 7, wherein the instructions, when executed by the processor, configure the processor to generate at least one third text prediction using a static language model, and wherein the processor is configured to combine the at least one first text prediction generated using the general language model, the at least one second text prediction from the context-specific language model and the at least one third text prediction from the static language model to generate a final set of text predictions.

9. The system of claim 8, wherein if a word candidate is generated by all language models the processor outputs the word candidate with a third associated probability, wherein the third associated probability is the greatest of a probability from the static language model and the product of the first and second probabilities from the general language model and the context-specific language model.

10. The system of claim 1, wherein at least one of the first text prediction or the second text prediction comprises a word candidate, and wherein when a word candidate generated using the general language model matches a word candidate generated using the context-specific language model, the processor is configured to calculate a linear combination of the first and second associated probability values generated by the general language model and the context-specific language model for that matching word.

11. The system of claim 10, wherein the processor calculates a weighted linear combination, wherein the weighting factor is based on a ratio of the second associated probability value of that word in the context-specific language model to the first associated probability value of that word in the general language model.

12. The system of claim 11, wherein the instructions, when executed by the processor, configure the processor to generate at least one third text prediction from a static language model, and combine the at least one first text prediction from the general language model, the at least one second text prediction from the context-specific language model and the at least one third text prediction from the static language model to generate a final set of text predictions.

13. The system of claim 12, wherein if a word candidate is generated by all language models the processor outputs the word candidate with a third associated probability, wherein the third associated probability is the greatest of a probability from the static language model and the weighted linear combination of the first and second associated probability values generated by the general language model and the context-specific language model.

14. The system of claim 1, wherein when the first term matches the second term, the step of combining comprises retaining the first prediction or the second prediction, dependent on whether the first associated probability value or the second associated probability value is highest.

15. The system of claim 1, wherein the context-specific language model comprises one of the group of:

an application-specific language model, a purpose or application-type specific language model, a company-specific language model, a user community-specific language model or a user language model.

16. A system comprising;

a processor; and

memory storing instructions that, when executed by the processor, configure the processor to;

receive text input by a user;

generate at least one first text prediction, each first text prediction comprising a first term and a first associated probability value, based on the text input by a user, using a first language model;

generate at least one second text prediction, each second text prediction comprising a second term and a second associated probability value, based on the text input by a user, using a second language model;

generate at least one weighted probability of the at least one first text prediction from the first language model using a first context-specific weighting factor, wherein the context-specific weighting factor is dependent on previous user'"'"'s use of that language model when inputting text for that context; and

generate final text predictions from the at least one weighted prediction generated from the first language model and the at least one second text prediction generated from the second language model.

17. The system of claim 16, wherein the instructions, when executed by the processor, further configure the processor to generate weighted probabilities of the at least one second text prediction from the second language model using a second context-specific weighting factor;

and

generate final text predictions from the weighted predictions generated from the first and second language models.

18. The system of claim 17, wherein the instructions, when executed by the processor, further configure the processor to generate text predictions using additional language models, each additional language model having an additional context-specific weighting factor associated with it.

19. The system of claim 18, wherein the system further comprises a data store having stored within it a set of context-specific parameters, the set of context-specific parameters comprising, for each context, a weighting factor for each language model.

20. The system of claim 19, wherein the context-specific parameters are recipient-specific parameters.

21. The system of claim 20, wherein the processor is configured to retrieve from the data store the recipient-specific weighting factors associated with a recipient to whom a message is being composed, and is configured to apply the respective recipient-specific weighting factor to the probabilities of the text predictions from each language model.

22. The system of claim 19, wherein the context-specific weighting factors are dynamic and are updated by the processor with use of the system.

23. A system comprising:

a processor; and

memory storing instructions that, when executed by the processor, configure the processor to;

receive text input by a user;

generate at least one first text prediction, each first text prediction comprising a first term and a first associated probability value, based on the text input by the user, using a general language model;

generate at least one second text prediction, each second text prediction comprising a second term and a second associated probability value, based on the text input by the user, using one of a plurality of recipient-specific dynamic language models, wherein each recipient-specific dynamic language model is trained on text sent to that recipient; and

combine the at least one first text prediction generated from the general language model and the at least one second text prediction from the one of the plurality of recipient-specific dynamic language models.

24. The system of claim 23, wherein the general language model is a general dynamic language model trained on user inputted text.

25. A non-transient computer readable medium containing program instructions which, when executed by a processor, configure the processor to:

generate at least one first text prediction, each first text prediction comprising a first term and a first associated probability value, based on the text input by the user, using a general language model;

generate at least one second text prediction, each second text prediction comprising a second term and a second associated probability value, based on the text input by the user, using one of a plurality of recipient-specific language models, wherein each recipient-specific language model is trained on text sent to that recipient; and

combine the at least one first text prediction generated from the general language model and the at least one second text prediction from the one of a plurality of recipient-specific language models.

26. A system comprising:

a processor; and

memory storing instructions that, when executed by the processor, configure the processor to;

receive text input by a user;

generate at least one first text prediction, each first text prediction comprising a term and a first associated probability value, based on the received text input, using a general language model;

generate at least one second text prediction, each second text prediction comprising a second term and a second associated probability value, based on the received text input, using a context-specific language model, wherein the context-specific language model is trained on user input text inputted from a specific location;

combine the at least one first text prediction generated using the general language model and the at least one second text prediction generated using the context-specific language model; and

output one or more of the combined text predictions.