Input recognition using multiple lexicons

• US 8,543,384 B2
• Filed: 09/14/2012
• Issued: 09/24/2013
• Est. Priority Date: 10/22/2005
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Kristensson, Per Ola, Zhai, Shumin
• Primary Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US13/616,311
• Publication Number:
  US 20130006639A1
• Time in Patent Office: 375 Days
• Field of Search: 704/1, 704/10, 704231-257
• US Class Current: 704/10
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 3/04883 for inputting data by handwriting, e.g. gesture or text
• CPC Class Codes: G06F 40/10 Text processing natural language analysis G06F40/20; semantic analysis G06F40/30; processing or translation of natural language G06F40/40
• CPC Class Codes: G06F 40/237 Lexical tools

Abstract

A word pattern recognition system improves text input entered via a shorthand-on-keyboard interface. A core lexicon comprises commonly used words in a language; an extended lexicon comprises words not included in the core lexicon. The system only directly outputs words from the core lexicon. Candidate words from the extended lexicon can be outputted and simultaneously admitted to the core lexicon upon user selection. A concatenation module enables a user to input parts of a long word separately. A compound word module combines two common shorter words whose concatenation forms a long word.

Claims

1. A word recognition system for recognizing an input signal entered by a user via a shorthand-on-keyboard interface, the system comprising:

a core lexicon comprising commonly used words,

wherein the commonly used words were selected for the core lexicon based on an associated frequency of use value for each word being above a pre-determined threshold, and

wherein the frequency of use values are not based on candidate selections by the user;

an extended lexicon comprising words not contained in the core lexicon;

a recognition module for recognizing words associated with the input signal;

a selector module for outputting an output word associated with the input signal from the core lexicon; and

a module for admitting, from the extended lexicon to the core lexicon, a candidate word associated with the input signal, upon a first selection of the candidate word by the user, to create an augmented core lexicon.

2. The system of claim 1, further comprising a user selection interface presenting candidate words associated with the input signal from at least one of the core lexicon and the extended lexicon, for selection by the user.

3. The system of claim 2, wherein the user selection interface lists candidate words from the core lexicon and candidate words from the extended lexicon with different perceptual features for ease of distinction.

4. The system of claim 1, wherein the recognition module generates an N-best list of candidate words from the core lexicon and the extended lexicon.

5. The system of claim 4, further comprising a pre-ranking module for ranking the N-best list of candidate words according to at least one criterion.

6. The system of claim 5, wherein the ranking module outputs a highest ranked word from the core lexicon as the highest ranked word in the N-best list of candidate words.

7. The system of claim 1, wherein the extended lexicon contains at least twice as many words as the core lexicon.

8. A word recognition method for recognizing an input text entered via a shorthand-on-keyboard interface, the method comprising:

storing commonly used words on a core lexicon,

wherein the commonly used words were selected for the core lexicon based on an associated frequency of use value for each word being above a pre-determined threshold, and

wherein the frequency of use values are not based on candidate selections by the user;

storing at least one word not contained in the core lexicon in an extended lexicon;

recognizing, via one or more processors, words associated with the input text;

outputting, via the one or more processors, an output word associated with the input text from the core lexicon; and

admitting, via the one or more processors, a candidate word associated with the input text, from the extended lexicon to the core lexicon, upon a first selection of the candidate word by a user, to create an augmented core lexicon.

9. The method of claim 8, further comprising presenting candidate words associated with the input text from at least one of the core lexicon and the extended lexicon, for selection by the user.

10. The method of claim 9, further comprising listing candidate words from the core lexicon and candidate words from the extended lexicon with different perceptual features for ease of distinction.

11. The method of claim 8, further comprising generating an N-best list of candidate words from the core lexicon and the extended lexicon.

12. The method of claim 11, further comprising ranking the N-best list of candidate words according to at least one criterion.

13. The method of claim 12, wherein the ranking includes outputting a highest ranked word from the core lexicon as the highest ranked word in the N-best list of candidate words.

14. The method of claim 8, wherein the extended lexicon contains at least twice as many words as the core lexicon.

15. A computer program product having program codes stored on a computer-readable storage device that, when executed by a computing device, causes the computing device to recognize an input signal entered via a user input interface, the program codes comprising:

a core lexicon comprising commonly used words,

wherein each of the commonly used words was selected for the core lexicon based on an associated frequency of use value for each word being above a pre-determined threshold, and

wherein the frequency of use values are not based on candidate selections by the user;

an extended lexicon comprising at least one word not contained in the core lexicon;

a program code that, when executed by the computing device, causes the computing device to recognize words associated with the input signal;

a program code that, when executed by the computing device, causes the computing device to output an output word associated with the input signal from the core lexicon; and

a program code that, when executed by the computing device, causes the computing device to admit a candidate word associated with the input signal, from the extended lexicon to the core lexicon, upon a first selection of the candidate word by the user, to create an augmented core lexicon.

16. The computer program product of claim 15, further comprising a program code that, when executed by the computing device, causes the computing device to present candidate words associated with the input signal from at least one of the core lexicon and the extended lexicon, for selection by the user, and wherein the extended lexicon contains at least twice as many words as the core lexicon.

17. The computer program product of claim 16, further comprising a program code that, when executed by the computing device, causes the computing device to list candidate words from the core lexicon and candidate words from the extended lexicon with different perceptual features for ease of distinction.

18. The computer program product of claim 15, further comprising a program code that, when executed by the computing device, causes the computing device to generate an N-best list of candidate words from the core lexicon and the extended lexicon.

19. The computer program product of claim 18, further comprising a program code that, when executed by the computing device, causes the computing device to rank the N-best list of candidate words according to at least one criterion.

20. The computer program product of claim 19, wherein the ranking includes outputting a highest ranked word from the core lexicon as the highest ranked word in the N-best list of candidate words.