IP Library › Granted Patent US 7,222,073
Granted Patent B2
US 7,222,073 · App. 09/999,515 · Granted May 22, 2007

System and method for speech activated navigation

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,222,073
App. No.
09/999,515
Granted
May 22, 2007
Kind
B2
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 (52)

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 database storing a list of content items organized by content title, each content title representing a database entry;

a speech recognizer transcribing an utterance into a digital signal representing a textual statement;

means for generating a grammar structure comprising various grammar paths, wherein every single word of each candidate content title that may be recognized is assigned to a single grammar path, said grammar structure further comprising a recursion loop by which every single word may be uttered any number of times and intermixed in order with any other word in said grammar structure;

means for recognizing the utterance by the speech recognizer that uses said grammar;

means for applying every single word of an output of said speech recognizer against each content title;

means for computing a matching score for each candidate content title;

means for mapping a content title with highest matching score into a command acceptable by a server which delivers the content item represented by said content title with highest matching score.

2. The system of claim 1 , wherein said matching score may be defined by one or more parameters to ensure reliability.

3. The system of claim 2 , wherein said one or more parameters are selected from a group consisting of:

number of times that a word appears in a content title;

relative position that a word appears in a content title;

relative order that a word appears in a content title;

length of candidate content title;

a fraction of words in an output of said speech processor that could match any word in a specific content title; or

said speech processor's confidence of each word.

4. The system of claim 1 , wherein said each candidate content title is selected from the whole database entries or a list of database entries that satisfy a specific criterion which excludes certain content titles.

5. A method for browsing a content database by spoken commands, said content database containing a list of content items organized by content title, each content title representing a database entry, said method comprising the steps of:

generating a grammar comprising various grammar paths, wherein every single word of each candidate content title that may be recognized is assigned to a single grammar path, said grammar further comprising a recursion loop by which every single word may be uttered any number of times and intermixed in order with any other word in said grammar structure;

recognizing the utterance by the speech recognizer that uses said grammar;

computing a matching score for each candidate content title; and

mapping a content title with highest matching score into a command acceptable by a server which delivers the content item represented by said content title with highest matching score.

6. The method of claim 5 , wherein said matching score is defined by multiple parameters.

7. The method of claim 5 , wherein said matching score is defined by one or more parameters selected from a group consisting of:

number of times that a word appears in a content title;

relative position that a word appears in a content title;

relative order that a word appears in a content tide;

length of candidate content title;

a fraction of words in an output of said speech recognition system that could match any word in a specific content title; or

said speech recognizer's confidence of each word.

8. The method of claim 5 , wherein said each candidate content title is selected from the entire database entries or a list of database entries that satisfy a specific criterion which excludes certain content titles.

9. A method for browsing a content database by spoken commands, said content database containing a list of content items organized by content title, each content title representing a database entry, said method comprising the steps of:

generating a grammar comprising various grammar paths, wherein every single word of a list of candidate content titles that may be recognized is assigned to a single grammar path, said grammar further comprising a recursion loop by which every single word may be uttered any number of times and intermixed in order with any other word in said grammar structure;

recognizing the utterance by the speech recognizer that uses said grammar;

matching each word of an output of a speech recognizer to said list of candidate content titles, said speech recognizer transcribing an audio signal into a digital signal representing said spoken commands;

computing a matching scare for each candidate content title; and

mapping a content title with highest matching score into a command acceptable by a server which delivers the content represented by said content title with highest matching score;

wherein said matching score is defined by one or more parameters selected from a group consisting of:

number of times that a word appears in a content title;

relative position that a word appears in a content title;

relative order that a word appears in a content title;

length of candidate content title;

a fraction of words in an output of said speech recognizer that could match any word in a specific content title; or

said speech recognizer's confidence of each word.

10. 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 database storing a list of content items organized by content title, each content title representing a database entry;

a speech recognizer transcribing an utterance into a digital signal representing a textual statement;

means for generating a grammar structure comprising various grammar paths, wherein every single word of each candidate content title that may be recognized is assigned to a distinct grammar path, said grammar structure further comprising a recursion loop by which every single word may be uttered any number of times and intermixed in order with any other word in said grammar structure;

means for recognizing the utterance by the speech recognizer that uses said grammar;

means for applying every single word of an output of said speech recognizer against each content title;

means for computing a matching scare for each candidate content title; and

means for mapping a content title with highest matching score into a command acceptable by a server which delivers the content item represented by said content title with highest matching score.