Method and Apparatus for Automatically Building Conversational Systems

• US 20100098224A1
• Filed: 12/22/2009
• Published: 04/22/2010
• Est. Priority Date: 12/19/2003
• Status: Active Grant

• Current Assignee: Interactions, LLC
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Feng, Junlan, Bangalore, Srinivas, Rahim, Mazin G.
• Granted Patent: US 8,175,230 B2
• Time in Patent Office: Days
• US Class Current: 379/88.17
• CPC Class Codes: G06F 40/174 Form filling; Merging
• CPC Class Codes: G06Q 10/06311 Scheduling, planning or task assignment for a person or group
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0635 Processing of requisition or of purchase orders
• CPC Class Codes: H04L 12/6418 Hybrid transport
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 2203/355 Interactive dialogue design tools, features or methods
• CPC Class Codes: H04M 3/4938 comprising a voice browser which renders and interprets, e.g. VoiceXML
• CPC Class Codes: H04M 7/0009 where voice calls remain entirely in the telephone network
• CPC Class Codes: H04M 7/0036 where the data service is an information service

Abstract

A system and method provides a natural language interface to world-wide web content. Either in advance or dynamically, webpage content is parsed using a parsing algorithm. A person using a telephone interface can provide speech information, which is converted to text and used to automatically fill in input fields on a webpage form. The form is then submitted to a database search and a response is generated. Information contained on the responsive webpage is extracted and converted to speech via a text-to-speech engine and communicated to the person.

Claims

1. (canceled)

2. In a natural language dialog, a method of providing information to a user utilizing worldwide web content, the method causing a computing device to perform steps comprising:

parsing a webpage according to fields in a web-based form on the webpage;

prompting a user to provide audible information associated with the parsed webpage;

receiving speech from the user in response to the prompt;

converting the speech to text;

automatically filling in a web-based form using the text converted from the speech;

receiving a dynamically generated webpage generated according to the filled webpage form; and

communicating to the user, via synthetic speech, content from the dynamically generated webpage.

3. The method of claim 2, wherein the speech relates to an entity name and location identifier.

4. The method of claim 3, wherein the entity name comprises a business name.

5. The method of claim 4, wherein the location identifier is an address.

6. The method of claim 5, wherein the location identifier is a zip code.

7. The method of claim 3, wherein the information provided to the user is location information.

8. The method of claim 7, wherein the information provided to the user comprises a listing of a set of the geographically closest entities to the location identifier.

9. A method of interacting with the worldwide web via a telephone, the method causing a computing device to perform steps comprising:

parsing a webpage to identify input fields in the webpage;

receiving speech from a user;

converting the speech to text;

filling in the webpage input fields for a database search based on the text;

extracting information from one or more dynamic webpage generated in response to the database search;

communicating a natural speech response to the user based on the extracted information.

10. A method of automatically generating a dialog interface to a website, the method causing a computing device to perform steps comprising:

parsing a website offline to identify input fields and website content;

generating prompts for use in a spoken dialog with a user associated with the parsed website;

receiving audible user input in response to at least one generated prompt;

automatically filling input fields on the website according to the received audible user input; and

presenting information to the user in response to the filled input field from the website.

11. The method of claim 10, wherein the step of parsing the website offline to identify input fields and website content only parses a portion of the website content, wherein at least a portion of the website parsing occurs dynamically during the spoken dialog.

12. The method of claim 10, wherein the step of generating prompts for use in a spoken dialog with a user associated with the website further utilizes previous interaction via the spoken dialog with the user.

13. A method of automatically generating a dialog interface to a website, the method causing a computing device to perform steps comprising:

(1) offline parsing of the website to identify input fields and website content;

(2) generating prompts using the offline parsed website for use in a spoken dialog with a user associated with the website;

(3) engaging in a dialog with the user utilizing the generated prompts; and

(4) if a webpage is dynamically generated from the website during the dialog;

(a) parsing the dynamically generated webpage;

(b) generating prompts associated with the dynamically generated webpage; and

(c) continuing the dialog with the user using either the dynamically generated prompts or the prompts generated from the offline parsed website content.

14. The method of claim 13, wherein the step of generating prompts for use in a spoken dialog with a user associated with the website further utilizes data from a previous dialog with the user.