System and method for inputting text into electronic devices

• US 10,146,765 B2
• Filed: 06/03/2016
• Issued: 12/04/2018
• Est. Priority Date: 09/29/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Touchtype Limited (Microsoft Corporation)
• Inventors: Medlock, Benjamin, Orr, Douglas Alexander Harper
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US15/173,412
• Publication Number:
  US 20160283464A1
• Time in Patent Office: 914 Days
• CPC Class Codes: G06F 3/0236 using selection techniques to select from displayed items
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs

Abstract

A text prediction engine, a system comprising a text prediction engine, and a method for generating sequence predictions. The text prediction engine, system and method generate multiple sequence predictions based on evidence sources and models, with each sequence prediction having a sequence and associated probability estimate.

Claims

1. A text prediction system, comprising:

one or more processors;

memory storing instructions that, when executed by the one or more processors, configure the one or more processors to;

generate a first sequence prediction based on a first evidence source including a first sequence of characters and a context of the first sequence of characters, and a first model that implements a context model including a first candidate model implementing a conditional distribution of the context for a first particular candidate and a prefix match model implementing a conditional distribution of candidates for a target sequence, wherein the first sequence prediction comprises a first sequence and a first associated probability estimate;

generate a second sequence prediction based on a second evidence source including a second sequence of characters and a second model that implements an input model including a second candidate model implementing a conditional distribution of inputs for a second particular candidate and a language model implementing a conditional distribution of sequences in a language for a particular context, wherein the second sequence prediction comprises a second sequence and a second associated probability estimate; and

a display coupled to at least one of the one or more processors or the memory, wherein the display is configured to;

output the first sequence prediction and the second sequence prediction within a text entry graphical user interface.

2. The system according to claim 1, wherein one of the first evidence source or the second evidence source is based upon input representing observed evidence about a word that a user is currently entering and the other of the first evidence source or the second evidence source is not based upon the input representing the observed evidence about the word the user is currently entering.

3. The system according to claim 2, wherein the first evidence source is modelled independently of the second evidence source.

4. The system according to claim 3, wherein the first model is different from the second model.

5. The system according to claim 4, wherein the first evidence source is modelled by the first model to generate the first sequence prediction and the second evidence source is modelled by the second model to generate the second sequence prediction.

6. The system according to claim 1, further comprising a prior model configured to generate a third set of sequences with associated probability estimates.

7. The system according to claim 6, wherein the prior model comprises a unigram model.

8. The system according to claim 6, wherein the prior model comprises a character model.

9. The system according to claim 1, wherein the first evidence source is modelled independently of the second evidence source.

10. The system according to claim 9, wherein the first evidence source is modelled by the first model to generate the first sequence prediction and the second evidence source is modelled by the second model to generate the second sequence prediction.

11. A computing device comprising one or more processors and memory storing instructions that, when executed by the one or more processors, configure the computing device to:

generate a first sequence prediction based on a first evidence source including a first sequence of characters and a context of the first sequence of characters, and a first model that implements a context model including a first candidate model implementing a conditional distribution of the context for a first particular candidate and a prefix match model implementing a conditional distribution of candidates for a target sequence, wherein the first sequence prediction comprises a first sequence and a first associated probability estimate;

generate a second sequence prediction based on a second evidence source including a second sequence of characters and a second model that implements an input model including a second candidate model implementing a conditional distribution of inputs for a second particular candidate and a language model implementing a conditional distribution of sequences in a language for a particular context, wherein the second sequence prediction comprises a second sequence and a second associated probability estimate; and

the computing device further comprising a display coupled to at least one of the one or more processors or the memory, wherein the display is configured to output the first sequence prediction and the second sequence prediction within a text entry graphical user interface.

12. A method for predicting text by a computing device, the method comprising:

generating, by the computing device, a first sequence prediction based on a first evidence source including a first sequence of characters and a context of the first sequence of characters, and a first model that implements a context model including a first candidate model implementing a conditional distribution of the context for a first particular candidate and a prefix match model implementing a conditional distribution of candidates for a target sequence, wherein the first sequence prediction comprises a first sequence and a first associated probability estimate;

generating, by the computing device, a second sequence prediction based on a second evidence source including a second sequence of characters and a second model that implements an input model including a second candidate model implementing a conditional distribution of inputs for a second particular candidate and a language model implementing a conditional distribution of sequences in a language for a particular context, wherein the second sequence prediction comprises a second sequence and a second associated probability estimate; and

outputting, on a display device communicatively coupled to the computing device, the first sequence prediction and the second sequence prediction within a text entry graphical user interface.