Natural expression processing method, processing and response method, device, and system

• US 9,753,914 B2
• Filed: 04/28/2016
• Issued: 09/05/2017
• Est. Priority Date: 10/28/2013
• Status: Expired due to Fees

• Current Assignee: Icontek Corporation
• Original Assignee: Zili Yu
• Inventors: Yu, Zili
• Primary Examiner(s): Chawan, Vijay B
• Application Number:
  US15/141,576
• Publication Number:
  US 20160253434A1
• Time in Patent Office: 495 Days
• Field of Search: 704 9, 704 2, 704257, 704 8, 704 4, 704 7, 704 10, 704231, 704251, 704254, 704270, 704275, 704 3
• CPC Class Codes: G06F 16/90332 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/42 Data-driven translation
• CPC Class Codes: G06F 40/47 Machine-assisted translation, e.g. using translation memory
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/008 based on physical entities controlled by simulated intelligence so as to replicate intelligent life forms, e.g. based on robots replicating pets or humans in their appearance or behaviour
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase

Abstract

A method involves: identifying a natural expression from a user and obtaining a certain form of language information which can be processed by a computer; and converting the identified and obtained language information to a standard expression in an encoded form. According to an example method, a natural expression is converted to an encoded standard expression; conversion to a standard expression is converting the semantics of a natural expression to encoding and parameters; precise verbatim translation is not necessary, thus the requirement for degree of accuracy of machine translation can be reduced; at the same time, the complexity of the database used for expression conversion (machine translation) is reduced, increasing data query and update speed and thus improving smart processing performance. Furthermore, the relatively simple encoded expression reduces the workload for manually-assisted interventions, increasing the efficiency of the work of manually-assisted interventions.

Claims

1. A computer-implemented method, comprising:

receiving, by a central controller computing device, a natural expression input, wherein the natural expression input is obtained via an interface of a client device and comprises a first form of language information;

identifying, by a robot computing device, a natural expression by converting the natural expression input to a second form of language information that can be processed by a computer; and

converting, by the robot computing device or manual aided understanding (MAU) workstation, the identified natural expression in the second form of language information to a standard expression in an encoded form, wherein converting the second form of language information to the standard expression comprises;

determining whether an understanding of the robot is mature based at least in part on an accuracy rate of the understanding of the robot over a certain time interval;

when it is determined that the understanding of the robot is mature, performing machine conversion to determine the standard expression;

when it is determined that the understanding of the robot is not mature the machine conversion, coordinating with the MAU workstation to perform manual conversion processing to determine the standard expression; and

outputting the standard expression of either the machine conversion or the manual conversion.

2. The computer-implemented method of claim 1, wherein, the standard expression includes requirement codes.

3. The computer-implemented method of claim 2, wherein, the requirement codes are represented by digital codes.

4. The computer-implemented method of claim 2, wherein, the standard expression further includes requirement parameters further embodying user-specified requirements.

5. The computer-implemented method of claim 4, wherein, the conversion of the natural expression input to the second form of language information is constituted by language information units obtained through spotting and conversion performed on the natural expression in the form of speech by using a modeling tool.

6. The computer-implemented method of claim 2, wherein, the conversion of the natural expression input to the second form of language information is constituted by language information units obtained through spotting and conversion performed on the natural expression in the form of speech by using a modeling tool.

7. The computer-implemented method of claim 1, wherein, the conversion of the natural expression input to the second form of language information is constituted by language information units obtained through spotting and conversion performed on the natural expression in the form of speech by using a modeling tool.

8. The computer-implemented method of claim 7, wherein, the second form of language information is constituted by one of phonemes, characters, and phrases.

9. The computer-implemented method of claim 7, wherein, conversion from the second form of language information to the standard expression is implemented on the basis of machine translation (MT) training dataset between the language information and the standard expression.

10. The computer-implemented method of claim 1, wherein, the second form of language information is constituted by one of phonemes, characters, and phrases.

11. The computer-implemented method of claim 1, wherein, conversion from the second form of language information to the standard expression is implemented on the basis of machine translation (MT) training dataset between the language information and the standard expression.

12. The computer-implemented method of claim 1, wherein, additional information associated with the natural expression is obtained during the identification of the natural expression, and the additional information is converted to a part of the standard expression.

13. The computer-implemented method of claim 1, further comprising:

wherein the identifying of the natural expression, is performed to further obtain relevant expression type information; and

wherein converting the natural expression in the second form of language information to the standard expression comprises;

determining whether the identified natural expression and the expression type information can be converted into an encoded standard expression through machine conversion

if it is determined that the desired standard expression cannot be obtained through the machine conversion, performing manual conversion processing;

invoking or generating a standard response matching with the standard expression of the machine conversion or the manual conversion; and

outputting the generated standard response in the manner of corresponding to the expression type information.

14. The computer-implemented method of claim 13, wherein, the standard response is fixed data pre-stored in a database, or the standard response is generated on the basis of variable parameters and basic data of the standard response pre-stored in the database.

15. A computing system comprising:

a dialogue gateway, a central controller, a manual aided understanding (MAU) workstation, a robot, an expression database, a response database, and a response generator;

wherein the dialogue gateway is configured to receive natural expression input obtained via an interface of a client device, to transmit the natural expression input to the central controller for subsequent processing, and to transmit a response to the natural expression input to the client device;

wherein the central controller is configured to receive the natural expression input from the dialogue gateway, to coordinate operation of the robot and the MAU workstation, to convert the natural expression input to a standard expression, and to instruct the response generator to generate a standard response corresponding to the standard expression;

wherein the robot is configured to, in response to the instruction of the central controller, process the natural expression input to identify the natural expression, by conversion of the natural expression input to language information that can be processed by a computer, and to convert the language information to the standard expression using the expression database;

wherein the MAU workstation is configured to present either the identified natural expression or the natural expression input via a manual-agent interface, to receive the standard expression in input data received via the manual-agent interface, and to transmit the standard expression to the central controller;

wherein the expression database is configured to store expression-related data comprising;

language information data associated with the natural expression, standard expression data associated with the standard expression, and data associated with a relationship between the language information and the standard expression;

wherein a response database stores response-related data, including standard response data for invocation, data for generating the response, or both; and

wherein the response generator is configured to receive instructions of the central controller, and to generate the response to the natural expression input using the data in the response database.

16. The computing system of claim 15, wherein, the central controller updates the expression database, the response database, or both.

17. The computing system of claim 15, further comprising a trainer device configured to train the robot to convert the natural expression into the standard expression.

18. The computing system of claim 15, wherein, the dialogue gateway further comprises an identity authenticator configured to identify and verify the user'"'"'s identity before receiving the natural expression, wherein authentication methods for the identification of a user at least include pass-phrase plus voice-print identification.