Platform for creating customizable dialog system engines

• US 9,275,641 B1
• Filed: 09/11/2015
• Issued: 03/01/2016
• Est. Priority Date: 09/14/2014
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Speaktoit, Inc. (Alphabet Inc.)
• Inventors: Gelfenbeyn, Ilya Genadevich, Goncharuk, Artem, Sirotin, Pavel Aleksandrovich
• Primary Examiner(s): He, Jialong
• Application Number:
  US14/851,145
• Publication Number:
  US 20160078866A1
• Time in Patent Office: 172 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04M 2203/355 Interactive dialogue design tools, features or methods
• CPC Class Codes: H04M 3/4217 Managing service interactions
• CPC Class Codes: H04M 3/4936 Speech interaction details speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/527 Centralised call answering arrangements not requiring operator intervention
• CPC Class Codes: H04M 7/0012 Details of application programming interfaces [API] for telephone networks; Arrangements which combine a telephonic communication equipment and a computer, i.e. computer telephony integration [CPI] arrangements

Abstract

Provided are systems and methods for creating custom dialog system engines. The system comprises a dialog system interface installed on a first server or a user device and a platform deployed on a second server. The platform is configured to receive dialog system entities and intents associated with a developer profile and associate the dialog system entities with the dialog system intents to form a custom dialog system engine associated with the dialog system interface. The web platform receives a user request from the dialog system interface, activates the custom dialog system engine based on identification, and retrieves the dialog system entities and intents. The user request is processed by applying the dialog system entities and intents to generate a response to the user request. The response is sent to the dialog system interface.

Claims

1. A method, comprising:

enabling a developer, by a first server comprising at least one processor and a memory storing processor-executable codes, to create a developer profile;

receiving, by the first server, one or more developer example requests, wherein each of the developer example requests is associated with one or more phrases;

determining, by the first server, one or more dialog system entities from the one or more example requests using a machine-learning technique, wherein the one or more dialog system entities are associated with the developer profile;

determining, by the first server, one or more dialog system intents from the one or more developer example requests using a machine-learning technique, wherein the one or more dialog system intents are associated with the developer profile;

associating, by the first server, the one or more dialog system entities with the one or more dialog system intents to form a custom dialog system engine;

linking, by the first server, the custom dialog system engine with a dialog system interface using the developer profile, wherein the dialog system interface is provided on a client user device or a web server;

receiving, by the first server or a second server, a user request from the dialog system interface, wherein the dialog system interface is installed on a user device or a third server;

identifying, by the first server or the second server, the dialog system interface based on the user request;

based on the identification of the dialog system interface, activating, by the first server or the second server, the custom dialog system engine and retrieving the one or more dialog system entities and the one or more dialog system intents;

processing, by the first server or the second server, the user request by applying the one or more dialog system entities and the one or more dialog system intents; and

generating, by the first server or the second server, a response to the user request based on the processing and sending the response to the dialog system interface.

2. The method of claim 1, wherein the one or more dialog system entities include one or more of the following:

a keyword and at least one synonym to the keyword, a keyword and at least one definition of the keyword, and a list of keywords defining objects of one class.

3. The method of claim 1, wherein the one or more dialog system intents include one or more of the following:

a dialog system interaction scheme, a rule defining a relation between a user request and a dialog system response, and a rule of relationship between a particular action and the one or more dialog system entities.

4. The method of claim 1, further comprising creating, by web tools of a platform interface, the one or more dialog system entities and the one or more dialog system intents based on a developer input.

5. The method of claim 1, wherein the identification is based on an identifier of the dialog system interface, the identifier accompanying the user request.

6. The method of claim 1, further comprising displaying the response to the user.

7. The method of claim 1, further comprising playing back the response to the user.

8. The method of claim 1, further comprising:

pre-processing the user request to recognize one or more spoken words, wherein the user request includes a voice input; and

transforming the voice input into text input.

9. A system comprising at least one processor and a memory storing processor-executable codes, wherein when the processor-executable codes are executed by the at least one processor, the processor-executable codes cause the at least one processor to:

receive, by the first server, one or more developer example requests, wherein each of the developer example requests is associated with one or more phrases;

determine one or more dialog system entities from the one or more developer example requests using a machine-learning technique, wherein the one or more dialog system entities are associated with a developer profile;

determine one or more dialog system intents from the one or more developer example requests using a machine-learning technique, wherein the one or more dialog system intents are associated with the developer profile;

associate the one or more dialog system entities with the one or more dialog system intents to form a custom dialog system engine;

link the custom dialog system engine with a dialog system interface using the developer profile, wherein the dialog system interface is provided on a client user device or a web server;

receive a user request from the dialog system interface;

identify the dialog system interface based on the user request;

based on the identification of the dialog system interface, activate the custom dialog system engine and retrieve the one or more dialog system entities and the one or more dialog system intents;

process the user request by applying the one or more dialog system entities and the one or more dialog system intents to the user request; and

generate a response to the user request based on the processing and send the response to the dialog system interface.

10. The system of claim 9, wherein the dialog system interface is further configured to enable the developer to create the developer profile.

11. The system of claim 9, wherein the user request is accompanied with an identifier of the dialog system interface for the identification.

12. The system of claim 9, wherein the processor-executable codes further cause the at least one processor to display the response to the user.

13. The system of claim 9, wherein the processor-executable codes further cause the at least one processor to playback the response to the user.

14. The system of claim 9, wherein the custom dialog system engine includes an Automatic Speech Recognizer (ASR), a Natural Language Processing (NLP), a dialog manager, and an output renderer.

15. The system of claim 14, wherein the ASR is configured to:

receive speech-based user inputs; and

process the speech-based user inputs into a sequence of parameter vectors.

16. The system of claim 14, wherein the ASR includes one or more of the following:

a pattern-based speech recognizer, a free-dictation recognizer, an address book based recognizer, and a dynamically created recognizer.

17. The system of claim 14, wherein the NLP is configured to:

disassemble recognized input received from the ASR;

parse the recognized input to produce utterances;

analyze the utterances utilizing one or more of the following;

morphological analysis, part-of-speech tagging, shallow parsing; and

map the recognized input or one or more parts of the recognized input to meaning representations.

18. The system of claim 14, wherein the dialog manager is configured to:

communicate with external resources;

process recognized input using one or more of the following;

discourse analysis, knowledge database query, and system action prediction based on discourse context; and

select an action based on the processing and analysis of the NLP.

19. The system of claim 14, wherein the output renderer is configured to:

transform action selected by the dialog manager into output, wherein the output includes an operation, a text message, an image, and a video message.

20. A non-transitory processor-readable medium having instructions stored thereon, which when executed by one or more processors, cause the one or more processors to implement a method, comprising:

receiving one or more developer example requests, wherein the developer example requests are associated with one or more phrases;

determining one or more dialog system entities from the one or more developer example requests using a machine-learning technique, wherein the one or more dialog system entities are associated with a developer profile;

determining one or more dialog system intents from the one or more developer example requests using a machine-learning technique, wherein the one or more dialog system intents are associated with the developer profile;

associating the one or more dialog system entities with the one or more dialog system intents to form a custom dialog system engine;

linking the custom dialog system engine with a dialog system interface using the developer profile, wherein the dialog system interface is provided on a client user device or a web server;

receiving a user request from the dialog system interface, wherein the dialog system interface is installed on a user device or a third server;

identifying the dialog system interface based on the user request;

based on the identification of the dialog system interface, activating the custom dialog system engine and retrieving the one or more dialog system entities and the one or more dialog system intents;

processing the user request by applying the one or more dialog system entities and the one or more dialog system intents; and

generating a response to the user request based on the processing and sending the response to the dialog system interface.