ENHANCED LINGUISTIC TRANSFORMATION

• US 20080077386A1
• Filed: 03/01/2007
• Published: 03/27/2008
• Est. Priority Date: 09/01/2006
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Gu, Liang, Gao, Yuqing, Zhang, Wei
• Granted Patent: US 7,881,928 B2
• Time in Patent Office: Days
• US Class Current: 704/3
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 21/06 Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids G10L15/26 takes precedence

Abstract

Techniques for enhanced linguistic transformation are disclosed. For example, a method of linguistic transformation includes the steps of providing at least one input to a plurality of modules, wherein at least one module has a different configuration than at least another module, obtaining at least one output from each of at least a subset of the plurality of modules, and generating a set of distinct outputs. The input and the output include linguistic representations and at least a portion of the output is a result of applying one or more linguistic transformations to at least a portion of the input. Techniques for displaying a plurality of results so as to emphasize component-level differences are also disclosed. By way of example, a method of displaying a plurality of results includes the steps of determining at least one primary result within the plurality of results; displaying the at least one primary result with at least a first set of color attributes; and displaying at least one alternative result with at least a second set of color attributes. Each result represents an application of at least one linguistic transformation to a common input and each result comprises a plurality of components.

Claims

1. A method of linguistic transformation, the method comprising the steps of:

providing at least one input to a plurality of modules, wherein at least one module has a different configuration than at least another module;

obtaining at least one output from each of at least a subset of the plurality of modules; and

generating an output set, wherein the output set comprises at least a portion of a set of distinct outputs obtained from the modules,

wherein the input and the output comprise linguistic representations and at least a portion of the output is a result of applying one or more linguistic transformations to at least a portion of the input.

2. The method of claim 1, further comprising the step of:

sorting the output set based on at least one score value associated with each member of the output set.

3. The method of claim 1, further comprising the steps of:

presenting the at least one output to a user; and

permitting the user to select at least one output from the presented outputs.

4. The method of claim 3, wherein the configuration of at least one module is modified based at least in part on the user selection.

5. The method of claim 3, wherein the manner in which future output is presented is altered based at least in part on the user selection.

10. The method of claim 1, further comprising the steps of:

determining at least one primary result within the output set;

displaying the at least one primary result with at least a first set of color attributes; and

displaying at least one alternative result with at least a second set of color attributes.

6. A method of speech recognition, the method comprising the steps of:

providing at least one spoken input to a plurality of speech recognition modules, wherein at least one speech recognition module has a different configuration than at least another speech recognition module;

obtaining at least one recognition output from each of at least a subset of the plurality of speech recognition modules; and

generating a recognition output set, wherein the recognition output set comprises at least a portion of a set of distinct recognition outputs obtained from the speech recognition modules.

7. The method of claim 6, further comprising the steps of:

permitting the user to select at least one recognition output from the recognition output set,

wherein at least one future recognition output set is altered based at least in part on at least one user selection.

8. The method of claim 6, further comprising the steps of:

providing the at least one recognition output to a plurality of translation modules, wherein at least one translation module has a different configuration than at least another translation module;

obtaining at least one translation output from each of at least a subset of the plurality of translation modules; and

generating a translation output set, wherein the translation output set comprises at least a portion of a set of distinct translation outputs obtained from the translation modules.

9. The method of claim 8, further comprising the steps of:

permitting the user to select at least one translation output from the translation output set,

wherein at least one future recognition output set is altered based at least in part on at least one user selection.

11. A system for linguistic transformation, the system comprising:

a memory; and

at least one processor coupled to the memory and operative to;

provide at least one input to a plurality of modules, wherein at least one module has a different configuration than at least another module;

obtain at least one output from each of at least a subset of the plurality of modules; and

generate an output set, wherein the output set comprises at least a portion of a set of distinct outputs obtained from the modules, and

wherein the input and the output comprise linguistic representations and at least a portion of the output is a result of applying one or more linguistic transformations to at least a portion of the input.

12. An article of manufacture for linguistic transformation, comprising a machine readable medium containing one or more programs which when executed implement the steps of:

providing at least one input to a plurality of modules, wherein at least one module has a different configuration than at least another module;

obtaining at least one output from each of at least a subset of the plurality of modules; and

generating an output set, wherein the output set comprises at least a portion of a set of distinct outputs obtained from the modules;

wherein the input and the output comprise linguistic representations and at least a portion of the output is the result of applying one or more linguistic transformations to at least a portion of the input.

13. A method of displaying a plurality of results so as to emphasize component-level differences, the method comprising the steps of:

determining at least one primary result within the plurality of results;

displaying the at least one primary result with at least a first set of color attributes; and

displaying at least one alternative result with at least a second set of color attributes,

wherein each result represents an application of at least one linguistic transformation to a common input, and

wherein each result comprises a plurality of components.

14. The method of claim 13, wherein the step of displaying at least one alternative result using at least one set of color attributes comprises:

displaying at least a portion of the components which are identical in the primary result and the alternative result with a first set of color attributes; and

displaying at least a portion of the components which are different in the primary result and the alternative result with a second set of color attributes.

15. The method of claim 13, wherein the step of displaying at least one alternative result using at least one set of color attributes comprises:

displaying at least a portion of the components which are present in both the primary result and the alternative result with a first set of color attributes;

displaying at least a portion of the components which are present in the primary result but not present in the alternative result with a second set of color attributes; and

displaying at least a portion of the components which are not present in the primary result but are present in the alternative result with a third set of color attributes.

16. The method of claim 13, wherein a set of values associated with at least one color attribute within at least one set of color attributes is used to indicate a level of difference between the primary result and the alternative result.

17. The method of claim 13, wherein the plurality of results is generated by performing the steps of:

providing at least one input to a plurality of modules, wherein at least one module has a different configuration than at least another module;

obtaining at least one output from each of at least a subset of the plurality of modules; and

generating a plurality of results, wherein the plurality of results comprises at least a portion of a set of distinct outputs obtained from the modules.

18. A system for displaying a plurality of results so as to emphasize component-level differences, the system comprising:

a memory; and

at least one processor coupled to the memory and operative to;

determine at least one primary result within the plurality of results;

display the at least one primary result with at least a first set of color attributes; and

display at least one alternative result with at least a second set of color attributes,

wherein each result represents an application of at least one linguistic transformation to a common input, and

wherein each result comprising a plurality of components.

19. An article of manufacture for displaying a plurality of results so as to emphasize component-level differences, the article comprising a machine readable medium containing one or more programs which when executed implement the steps of:

determining at least one primary result within the plurality of results;

displaying the at least one primary result with at least a first set of color attributes; and

displaying at least one alternative result with at least a second set of color attributes,

wherein each result represents an application of at least one linguistic transformation to a common input, and

wherein each result comprises a plurality of components.

20. A method of linguistic transformation, the method comprising the steps of:

providing at least one input to a plurality of modules, wherein at least one module has a different configuration than at least another module;

obtaining at least one output from each of at least a subset of the plurality of modules;

generating an output set, wherein the output set comprises at least a portion of a set of distinct outputs obtained from the modules,

determining at least one primary result within the output set;

displaying the at least one primary result with at least a first set of color attributes; and

displaying at least one alternative result with at least a second set of color attributes,

wherein the input and the output comprise linguistic representations and at least a portion of the output is a result of applying one or more linguistic transformations to at least a portion of the input.