System Apparatus Circuit Method and Associated Computer Executable Code for Natural Language Understanding and Semantic Content Discovery

• US 20140236572A1
• Filed: 02/20/2013
• Published: 08/21/2014
• Est. Priority Date: 02/20/2013
• Status: Active Grant

• Current Assignee: Jinni Media Limited
• Original Assignee: Jinni Media Limited
• Inventors: Meshulam, Ram, Rimon, Mordecai, Ben-Zaken, Izhak
• Granted Patent: US 9,123,335 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G01L 17/00 Devices or apparatus for measuring tyre pressure or the pressure in other inflated bodies
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 2015/227 of the speaker; Human-factor methodology
• CPC Class Codes: G10L 25/54 for retrieval

Abstract

Disclosed are systems, apparatuses, circuits and methods for extrapolating meaning from vocalized speech or otherwise obtained text. Speech of a speaking user is sampled and digitized, the digitized speech is converted into a text stream, the text stream derived from speech or otherwise obtained is analyzed syntactically and semantically, a knowledgebase in the specific context domain of the text stream is utilized to construct one or more semantic/syntactic domain specific query analysis constrains/rule-sets, and a “Domain Specific Knowledgebase Query” (DSKQ) or set of queries is built at least partially based on the domain specific query analysis constrains/rule-sets.

Claims

1. A language understanding system for extrapolating meaning from vocalized speech comprising:

a Speech Acquisition Block to sample and digitize vocalized speech of a speaking user;

a Speech Analytics Block to analyze the digitized speech so as to recognize within the digitized speech samples a specific text stream;

a Domain Block to utilize knowledgebase, in a specific context domain of the vocalized speech, to construct one or more semantic/syntactic domain specific query analysis rule-sets; and

a Natural Language Processing Block to build, at least partially based on the domain specific query analysis rule-sets, a “

Domain Specific Knowledgebase Query”

(DSKQ) or set of queries.

2. The language understanding system of claim 1 wherein:

the Domain Block is further adapted to utilize knowledgebase, in the context domain of the vocalized speech, to construct one or more semantic domain specific content discovery rule-sets; and

, wherein the system further comprises;

a Data Discovery Block to find and output content(s) at least partially based on;

(a) the DSKQ(s), and/or (b) the semantic domain specific discovery rule-sets.

3. The language understanding system of claim 2 wherein:

the Speech Analytics Block further comprises a Voice Recognition Module to identify the speaking user by referencing a User Specific Data Block with digitized speech samples from the vocalized speech and relaying the identification data to the Data Discovery Block, and wherein;

said Data Discovery Block, as part of finding and outputting content(s), takes into consideration prior preferences of the speaking user by referencing said User Specific Data Block with the details of the identified speaking user.

4. The language understanding system of claim 2 wherein:

the Speech Analytics Block further comprises a Tempo-Spectral Analysis Module to derive Prosody information to assist the Natural Language Processing Block in parsing the digitized speech, as part of building the DSKQ(s).

5. The language understanding system of claim 4 wherein:

the Tempo-Spectral Analysis Module is further adapted to embed metadata indicative of the derived Prosody information into the specific text stream recognized by the Speech Analytics Block in the digitized speech, prior to relaying the now enhanced text stream to the Natural Language Processing Block.

6. The language understanding system of claim 2 wherein:

the Domain Block comprises multiple Domain Specific Knowledgebase Modules, based on a semantic taxonomy also referred to as Genome.

7. The language understanding system of claim 6 wherein:

one or more specific Domain Specific Knowledgebase Modules, out of the multiple Domain Specific Knowledgebase Modules in the Domain Block, are selected based on an explicit user request.

8. The language understanding system of claim 6 wherein:

a specific Domain Specific Knowledgebase Module, out of the multiple Domain Specific Knowledgebase Modules in the Domain Block, is derived from a general user request.

9. The language understanding system of claim 2 wherein:

the Data Discovery Block, upon reaching an equivocal query or content discovery result, presents to the speaking user over a functionally associated computerized device, the temporary results reached.

10. The language understanding system of claim 9 wherein:

the Data Discovery Block further presents to the speaking user with request to provide further details about his sought after content, and content is selected based on both;

(a) the analysis of the first text stream, and (b) the analysis of one or more additional text stream(s) provided by the speaking user in response to the presented request and wherein at least some of the one or more additional text stream(s) provide anaphoric references to terms introduced in prior text stream(s).

11. A language processing system comprising:

a natural language processing module designed to apply a domain specific rule set, expressed as a formal grammar, to a text stream;

a grammar parsing logic with a capability for dynamic loading or generation of lexical entries and meta-data variables and recovery of partial structures from failed or incomplete parses; and

a query analysis logic adapted to convert an output of said natural language processing module into a structured query, using a domain specific genome and logic for resolution of ambiguities.

12. A language understanding system for extrapolating meaning from natural language comprising:

a Domain Block to utilize knowledgebase, in the specific context domain of the natural language, to construct one or more semantic/syntactic domain specific query analysis rule-sets; and

a Natural Language Processing Block to build, at least partially based on the domain specific query analysis rule-sets, a “

Domain Specific Knowledgebase Query”

(DSKQ) or set of queries.

13. The language understanding system of claim 12 wherein:

the Domain Block is further adapted to utilize knowledgebase, in the context domain of the natural language, to construct one or more semantic domain specific discovery rule-sets; and

, wherein the system further comprises;

a Data Discovery Block to find and output content(s) at least partially based on;

(a) the DSKQ(s), and/or (b) the semantic domain specific discovery rule-sets.

14. The language understanding system of claim 12 wherein:

the Domain Block comprises multiple Domain Specific Knowledgebase Modules, based on a semantic taxonomy also referred to as Genome.

15. The language understanding system of claim 14 wherein:

one or more specific Domain Specific Knowledgebase Modules, out of the multiple Domain Specific Knowledgebase Modules in the Domain Block, are selected based on an explicit user request.

16. The language understanding system of claim 14 wherein:

a specific Domain Specific Knowledgebase Module, out of the multiple Domain Specific Knowledgebase Modules in the Domain Block, is derived from a general user request.

17. The language understanding system of claim 13 wherein:

the Data Discovery Block, upon reaching an equivocal query or content discovery result, presents to the speaking user over a functionally associated computerized device, the temporary results reached.

18. The language understanding system of claim 17 wherein:

the Data Discovery Block further presents to the speaking user with request to provide (i.e. by further speech) further details about his sought after content.

19. A method for language understanding and extrapolating meaning from natural language comprising:

utilizing knowledgebase, in the specific context domain of the natural language, to construct one or more semantic/syntactic domain specific query analysis rule-sets; and

building, at least partially based on the domain specific query analysis rule-sets, a “

Domain Specific Knowledgebase Query”

(DSKQ) or set of queries.

20. The method of claim 19 further comprising:

utilizing knowledgebase, in the context domain of the natural language, to construct one or more semantic domain specific discovery rule-sets;

finding content(s) at least partially based on;

(a) the DSKQ(s), and/or (b) the semantic domain specific discovery/recommendation rule-sets; and

outputting the found content(s).