Ordering internet voice content according to content density and semantic matching

• US 7,873,900 B2
• Filed: 09/09/2005
• Issued: 01/18/2011
• Est. Priority Date: 03/22/2002
• Status: Active Grant

• Current Assignee: S Aqua Semiconductor LLC (Intellectual Ventures LLC)
• Original Assignee: Inet Spch Property Hldg LLC (Intellectual Ventures LLC)
• Inventors: Khan, Emdadur R.
• Primary Examiner(s): Paula; Cesar B
• Application Number:
  US11/223,281
• Publication Number:
  US 20060010386A1
• Time in Patent Office: 1,957 Days
• Field of Search: 379/88.01, 379/88.17, 715/500, 715/531, 715/513, 715/200, 715/205, 715/234, 715/206, 715/501.1, 434/167, 434/321, 434/322, 434/362
• US Class Current: 715/234
• CPC Class Codes: G06F 16/9577 Optimising the visualization of content, e.g. distillation of HTML documents

Abstract

A method of establishing a bi-directional voice communication link with a network access provider, locating a page responsive to a voice command, identifying one or more highlights on the page, transmitting via the bi-directional voice communication link, a voice response corresponding to a aural recitation of the one or more highlights, receiving a voice selection signal identifying a selected one of the one or more highlights, and locating content associated with the selected one of the one or more highlights.

Claims

1. A method, comprising:

establishing a bi-directional voice communication link with an audio network access provider;

locating a network page responsive to a voice command signal corresponding to a network command using an intelligent agent configured to read a network content originally written for a visual web browser;

identifying one or more highlights associated with the network page in the network content originally written for the visual web browser comprising identifying related content on the network page and locating the related content;

transmitting, via the bi-directional voice communication link, a voice response signal corresponding to a network data signal representing a recitation of only the one or more highlights;

receiving, via the bi-directional voice communication link, a voice selection signal identifying a selected one of the one or more highlights in response to the voice response signal corresponding to the network data signal representing the recitation of only the one or more highlights;

locating network content originally written for the visual web browser identified by the selected one of the recited one or more highlights responsive to the voice selection signal;

separating vertically distinct pluralities of text within the identified network content originally written for the visual web browser responsive to the voice selection signal;

determining a content density associated with the related content;

determining multiple related contents responsive to the content density;

dividing the multiple related contents into one or more subgroups responsive to the content density;

determining semantic matching between at least two subgroups;

assigning weighted scores to the at least two subgroups responsive to the determining; and

ordering the at least two subgroups according to the assigning.

2. The method of accessing the network of claim 1, where the content originally written for the visual web browser includes content written using an hypertext markup language (HTML) page.

3. The method of claim 1 where identifying the one or more highlights includes:

identifying related content for each highlight on the network page;

locating the related content for each highlight;

determining an extent of the related content for each highlight; and

prioritizing the one or more highlights for serially ordered recitation according to the extent of the related content of each highlight.

4. The method of claim 1 where identifying the one or more highlights includes determining one or more words associated with the one or more highlights on the network page that match one or more words on a related network page and ordering the one or more highlights for serially ordered recitation according to the size of contents associated with the one or more words associated with the one or more highlights.

5. The method of claim 4 comprising:

determining multiple words on the related network page that match at least one word on the network page; and

ordering the multiple words according to an associated content size.

6. The method of claim 1 wherein the recitation is free of any items added to the content available on the network page.

7. A method, comprising:

establishing a bi-directional voice communication link with an audio network access provider;

locating a network page responsive to a voice command signal corresponding to a network command using an intelligent agent configured to read a network content originally written for a visual web browser;

identifying one or more highlights associated with the network page in the network content originally written for the visual web browser comprising identifying related content on the network page and locating the related content;

transmitting, via the bi-directional voice communication link, a voice response signal corresponding to a network data signal representing a recitation of only the one or more highlights;

receiving, via the bi-directional voice communication link, a voice selection signal identifying a selected one of the one or more highlights in response to the voice response signal corresponding to the network data signal representing the recitation of only the one or more highlights;

locating network content originally written for the visual web browser identified by the selected one of the recited one or more highlights responsive to the voice selection signal;

separating vertically distinct pluralities of text within the identified network content originally written for the visual web browser responsive to the voice selection signal;

determining a content density associated with the related content;

determining multiple related contents responsive to the content density;

dividing the multiple related contents into one or more subgroups responsive to the content density;

identifying one or more paragraphs associated with the one or more subgroups;

generating summaries of the one or more paragraphs;

determining semantic matching between at least two paragraphs; and

ordering the related content according to the semantic matching between the at least two paragraphs.

8. The method of accessing the network of claim 7, where the content originally written for the visual web browser includes content written using an hypertext markup language (HTML) page.

9. The method of claim 7 where receiving the voice response signal includes determining one or more words associated with the one or more highlights on the network page that match one or more words on a related network page and ordering the one or more highlights for serially ordered recitation according to the size of contents associated with the one or more words.

10. The method of claim 9 comprising:

determining multiple words on the related network page that match one word on the network page; and

ordering the multiple words according to an associated content size.

11. A method comprising:

establishing a bi-directional voice communication link with an audio Internet access provider;

initiating access to an Internet page corresponding to an Internet surfing command by transmitting, via the bi-directional voice communication link using an intelligent agent for reading an Internet content originally written for a visual web browser from the Internet page, a voice command signal corresponding to the Internet surfing command;

receiving, via the bi-directional voice communication link, a voice response signal corresponding to an Internet data signal representing a recitation of one or more highlights identified as being associated with the Internet content originally written for the visual web browser;

initiating access to Internet content originally written for the visual web browser related to a selected one of the recited one or more highlights by transmitting, via the bi-directional voice communication link, a voice selection signal identifying the selected one of the one or more highlights responsive to the voice response signal;

determining whether one or more words associated with the one or more highlights on the Internet page match one or more words on a related Internet page;

identifying related content on the Internet page;

locating the related content;

determining a content density associated with the related content;

determining multiple related contents responsive to the content density;

dividing the multiple related contents into one or more subgroups responsive to the content density;

determining semantic matching between at least two of the subgroups;

assigning weighted scores to the at least two subgroups responsive to the determining; and

ordering the at least two subgroups according to the assigning.

12. A method comprising:

establishing a bi-directional voice communication link with an audio Internet access provider;

initiating access to an Internet page corresponding to an Internet surfing command by transmitting, via the bi-directional voice communication link using an intelligent agent for reading an Internet content originally written for a visual web browser from the Internet page, a voice command signal corresponding to the Internet surfing command;

receiving, via the bi-directional voice communication link, a voice response signal corresponding to an Internet data signal representing a recitation of one or more highlights identified as being associated with the Internet content originally written for the visual web browser;

initiating access to Internet content originally written for the visual web browser related to a selected one of the recited one or more highlights by transmitting, via the bi-directional voice communication link, a voice selection signal identifying the selected one of the one or more highlights responsive to the voice response signal;

determining whether one or more words associated with the one or more highlights on the Internet page match one or more words on a related Internet page;

identifying related content on the Internet page;

locating the related content;

determining a content density associated with the related content;

determining multiple related contents responsive to the content density;

dividing the multiple related contents into one or more subgroups responsive to the content density;

identifying one or more paragraphs associated with the one or more subgroups;

generating summaries of the one or more paragraphs;

determining semantic matching between at least two paragraphs; and

ordering the related content according to the semantic matching between the at least two paragraphs.

13. A computer readable medium having computer-executable instructions stored thereon that, if executed by a processor, cause the processor to perform a method comprising:

establishing a bi-directional voice communication link with an audio network access provider responsive to a voice command corresponding to a network surfing command;

locating a network page corresponding to the network surfing command using an intelligent agent configured to read network content originally written for a visual web browser;

identifying one or more highlights associated with the network page in the network content originally written for the visual web browser comprising identifying related content on the network page and locating the related content;

transmitting, via the bi-directional voice communication link, a voice response signal corresponding to a network data signal representing a recitation of the one or more highlights;

receiving, via the bi-directional voice communication link, a voice selection signal identifying a selected one of the one or more highlights in response to the voice response signal corresponding to the network data signal representing the recitation of the one or more highlights;

locating network content originally written for the visual web browser selected one of the recited one or more highlights responsive to the voice selection signal;

determining a content density associated with the related content;

determining multiple related contents responsive to the content density;

dividing the multiple related contents into one or more subgroups responsive to the content density;

determining semantic matching between at least two subgroups;

assigning weighted scores to the at least two subgroups responsive to the determining; and

ordering the at least two subgroups according to the assigning.

14. The computer readable medium of claim 13 where identifying the one or more highlights includes determining one or more words associated with the one or more highlights on the network page that match one or more words on a related network page and ordering the one or more highlights for serially ordered recitation according to the size of contents associated with the one or more words.

15. The computer readable medium of claim 14 comprising:

determining multiple words on the related network page that match at least one word on the network page; and

ordering the multiple words according to an associated content size.

16. A computer readable medium having computer-executable instructions stored thereon that, if executed by a processor, cause the processor to perform a method comprising:

establishing a bi-directional voice communication link with an audio a network access provider responsive to a voice command corresponding to a network surfing command;

locating a network page corresponding to the network surfing command using an intelligent agent configured to read a network content originally written for a visual web browser;

identifying one or more highlights associated with the network page in the network content originally written for the visual web browser comprising identifying related content on the network page and locating the related content;

transmitting, via the bi-directional voice communication link, a voice response signal corresponding to a network data signal representing a recitation of the one or more highlights;

receiving, via the bi-directional voice communication link, a voice selection signal identifying a selected one of the one or more highlights in response to the voice response signal corresponding to the network data signal representing the recitation of the one or more highlights;

locating network content originally written for the visual web browser selected one of the recited one or more highlights responsive to the voice selection signal;

determining a content density associated with the related content;

determining multiple related contents responsive to the content density;

dividing the multiple related contents into one or more subgroups responsive to the content density;

identifying one or more paragraphs associated with the one or more subgroups;

generating summaries of the one or more paragraphs;

determining semantic matching between at least two paragraphs; and

ordering the related content according to the semantic matching between the at least two paragraphs.

17. An apparatus, comprising:

means for establishing a voice link with an access provider;

means for locating a page comprising content written for a visual web browser responsive to a voice command;

means for identifying one or more highlights on the page;

means for locating content related to the one or more highlights on the page;

means for transmitting, via the voice link, a voice response corresponding to an aural recitation of the one or more highlights;

means for receiving, via the voice link, a voice selection indicating selection of one of the one or more highlights responsive to the voice response;

means for locating additional content written for the visual web browser responsive to the selected one of the one more or more highlights;

means for determining a content density associated with the additional content;

means for determining multiple related content responsive to the content density;

means for determining semantic matching between at least two subgroups of the multiple related content responsive to the content density;

means for assigning weighted scores to the at least two subgroups responsive to the semantic matching; and

means for ordering the at least two subgroups according to the weighted scores.

18. An apparatus, comprising:

means for establishing a voice link with an access provider;

means for locating a page comprising content written for a visual web browser responsive to a voice command;

means for identifying one or more highlights on the page;

means for locating content related to the one or more highlights on the page;

means for transmitting, via the voice link, a voice response corresponding to an aural recitation of the one or more highlights;

means for receiving, via the voice link, a voice selection indicating selection of one of the one or more highlights responsive to the voice response;

means for locating additional content written for the visual web browser responsive to the selected one of the one more or more highlights;

means for determining a content density associated with the additional content;

means for determining multiple related content responsive to the content density;

means for identifying one or more paragraphs associated with one or more subgroups of the multiple related content responsive to the content density;

means for generating summaries of the one or more paragraphs;

means for determining semantic matching between at least two paragraphs; and

means for ordering the related content according to the semantic matching between the at least two paragraphs.

19. An apparatus, comprising:

a transaction processing unit; and

an understanding unit configured to operate together with the transaction processing unit to;

establish a voice link with an access provider and locate a page comprising content written for a visual web browser responsive to a voice command;

identify one or more highlights on the page and locate content related to the one or more highlights on the page;

transmit, via the voice link, a voice response corresponding to an aural recitation of the one or more highlights;

receive, via the voice link, a voice selection indicating selection of one of the one or more highlights responsive to the voice response;

locate additional content written for the visual web browser responsive to the selected one of the one more or more highlights;

determine a content density associated with the additional content;

determine multiple related content responsive to the content density;

identify one or more paragraphs associated with one or more subgroups of the multiple related content responsive to the content density;

generate summaries of the one or more paragraphs;

determine semantic matching between at least two paragraphs; and

order the related content according to the semantic matching between the at least two paragraphs.

20. An apparatus, comprising:

a transaction processing unit configured to establish a voice link with an access provider and locate a page comprising content written for a visual web browser responsive to a voice command;

an understanding unit configured to identify one or more highlights on the page and locate content related to the one or more highlights on the page;

where the transaction processing unit is further configured to;

transmit, via the voice link, a voice response corresponding to an aural recitation of the one or more highlights;

receive, via the voice link, a voice selection indicating selection of one of the one or more highlights responsive to the voice response; and

locate additional content written for the visual web browser responsive to the selected one of the one more or more highlights; and

where the understanding unit is further configured to;

determine a content density associated with the additional content;

determine multiple related content responsive to the content density;

identify one or more paragraphs associated with one or more subgroups of the multiple related content responsive to the content density;

generate summaries of the one or more paragraphs;

determine semantic matching between at least two paragraphs; and

order the related content according to the semantic matching between the at least two paragraphs.

21. A method, comprising:

establishing a voice link with an access provider;

issuing a voice command to locate a page comprising content written for a visual web browser responsive to a voice command;

receiving a voice response, via the voice link, corresponding to an aural recitation of one or more highlights identified on the page;

issuing a voice selection indicating selection of the one or more highlights responsive to the voice response;

receiving an aural recitation of an ordered list of at least two subgroups of multiple related content according to corresponding weighted scores;

wherein additional content written for the visual web browser is located responsive to the selected one or more highlights;

wherein the multiple related content is determined responsive to a content density associated with the additional content;

wherein the at least two subgroups have a semantic matching determined responsive to the content density; and

wherein the weighted scores are assigned to the at least two subgroups according to the semantic matching.

22. A method, comprising:

establishing a voice link with an access provider;

issuing a voice command to locate a page comprising content written for a visual web browser responsive to a voice command;

receiving a voice response, via the voice link, corresponding to an aural recitation of one or more highlights identified on the page;

issuing a voice selection indicating selection of the one or more highlights responsive to the voice response;

receiving an aural recitation of an ordered list of at least two paragraphs in multiple related content according to corresponding weighted scores;

wherein additional content written for the visual web browser is located responsive to the selected one or more highlights;

wherein the multiple related content is determined responsive to a content density associated with the additional content;

wherein the at least two paragraphs are identified in the multiple related content;

wherein the at least two paragraphs have a semantic matching determined responsive to the content density; and

wherein the weighted scores are assigned to the at least two paragraphs according to the semantic matching.