System and method for speech activated navigation

• US 20050055218A1
• Filed: 10/22/2004
• Published: 03/10/2005
• Est. Priority Date: 10/24/2001
• Status: Abandoned Application

• Current Assignee: Jehan G. Bing, Jerome Dubreuil, Luc E. Julia
• Original Assignee: Jehan G. Bing, Jerome Dubreuil, Luc E. Julia
• Inventors: Dubreuil, Jerome, Julia, Luc E., Bing, Jehan G.
• Application Number:
  US10/971,171
• Publication Number:
  US 20050055218A1
• Time in Patent Office: Days
• US Class Current: 704/277
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

The invention discloses a system and method for speech-activated navigating or browsing via a speech control interface used in a speech-activated multifunctional communications system. In one embodiment, the invention provides an approach to extend speech-activated navigation by linking an output of an open vocabulary recognizer to an Internet search engine in order that a user may have more options to search information related to his spoken commands. In another embodiment, the invention provides a means to enable the user to orally navigate a database via a speech control interface wherein the selections and associated selection criteria are organized into a hierarchical view menu. In another embodiment, the invention provides an approach with high flexibility and accuracy to recognize the user'"'"'s command using a new grammar structure and a matching score system.

Claims

1. A computer readable storage medium encoded with instructions, which when loaded into a digital computational device establishes a speech-activated browsing system, the system comprising:

a speech recognition system which is used to recognize input utterance against a list of selections;

means for dynamically generating a set of grammars from a hierarchical menu showing said selections and associated selection criteria, each of said grammar reflecting a selection criterion;

a database embodying the same list of selections and the associated selection criteria present in said hierarchical menu, each selection representing an entry in said database, and each entry being associated with a selection criterion that said entry satisfies;

means for applying a particular selection criterion against all entries of said database; and

means for returning a list of database entries that satisfy said particular selection criterion.

2. The system of claim 1, wherein said grammars for each different selection criterion are embedded in a larger grammar or a set of grammars which is constructed in such a manner that either a sequence of spoken commands or a fluent statement may be referenced to all database entries with desired property.

3. The system of claim 2, wherein said sequence of spoken commands or said fluent statement may be referenced to a list of candidate selections if said sequence of spoken commands or said fluent statement does not exactly match a certain selection criterion.

4. In a speech-activated communications system having a speech recognition unit, a central processing unit, and a database containing a list of items, said list of items being represented by a hierarchical menu, each item in said database being referenced to a selection and an associated selection criterion present in said hierarchical menu, a speech-activated browsing system comprising:

means for generating one or more grammars, reflecting different selection criteria which may be applied against entries of said database,

means for applying a particular selection criterion against all entries of said database; and

means for returning a list of database entries that satisfy said particular selection criterion.

wherein said grammars for each different selection criterion are embedded in a larger grammar or a set of grammars which is constructed in such a manner that either a sequence of spoken commands or a fluent statement may be referenced to a set of database entries with desired property; and

wherein said sequence of spoken commands or said fluent statement may be referenced to a list of candidate selections if said sequence of spoken commands or said fluent statement does not match a certain selection criterion.

5. A method for browsing a hierarchical menu by spoken commands, wherein said hierarchical menu contains a list of selections with associated selection criteria, said method comprising the steps of:

entering a set of selection criteria by an input utterance;

generating a set of grammars, each of said grammar reflecting a specific selection criterion;

applying said grammars against all entries of a database embodying the same list of selections and the same selection criteria present in said hierarchical menu, wherein each selection represents an entry in said database, and wherein each entry is associated with a selection criterion that said entry satisfies;

returning a list of entries that satisfy said selection criteria;

wherein said grammars for each different selection criterion are embedded in a larger grammar or a set of grammars which is constructed in such a manner that either a sequence of spoken commands or a fluent statement may be referenced to a set of database entries with desired property; and

wherein said sequence of spoken commands or said fluent statement may be referenced into a set of candidate selections if said sequence of spoken commands or said fluent statement does not match a certain selection criterion.

6. A method for browsing a hierarchical menu by spoken commands, wherein said hierarchical menu contains a list of selections with associated selection criteria, said method comprising the steps of:

entering a sequence of selection criteria by an input utterance;

generating a set of grammars, each of said grammar reflecting a specific selection criterion;

applying a first grammar against all entries of a database embodying the same list of selections and the same selection criteria present in said hierarchical menu, each selection representing an entry in said database, and each entry being associated with a selection criterion that said entry satisfies;

applying a second grammar to the database entries that satisfy the first selection criterion represented by said first grammar;

applying a third grammar to the database entries that satisfy the second selection criterion represented by said second grammar;

repeating the steps of applying grammars until the last grammar is applied;

returning a list of entries that satisfy said selection criteria.

wherein said grammars for each different selection criterion are embedded in a larger grammar or a set of grammars which is constructed in such a manner that either a sequence of spoken commands or a fluent statement may be referenced to a set of database entries with desired property.

7. The method of claim 6, wherein said sequence of spoken commands or said fluent statement may be referenced to a list of candidate selections if said sequence of spoken commands or said fluent statement does not match a certain selection criterion.