Using voice-based web navigation to conserve cellular data

• US 10,083,004 B2
• Filed: 09/24/2015
• Issued: 09/25/2018
• Est. Priority Date: 12/18/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Clark, Adam T., Huebert, Jeffrey K., Payton, Aspen L., Petri, John E.
• Primary Examiner(s): ADESANYA, OLUJIMI A
• Application Number:
  US14/863,855
• Publication Number:
  US 20160179465A1
• Time in Patent Office: 1,097 Days
• Field of Search: 704235, 704246, 704270-275
• CPC Class Codes: G06F 16/9566 URL specific, e.g. using aliases, detecting broken or misspelled links
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G06F 40/00 Handling natural language data speech analysis or synthesis, speech recognition G10L
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers
• CPC Class Codes: H04W 4/16 Communication-related supplementary services, e.g. call-transfer or call-hold

Abstract

Techniques disclose methods for providing voice-based web navigation through a cellular voice connection. A voice command a request to return content of a webpage having one or more topics is received via a cellular voice connection. One or more of the topics of the webpage that match the content specified in the request is identified. An audio indication of the content is returned via the cellular voice connection.

Claims

1. A method comprising:

receiving, via a cellular voice connection initiated by a mobile device, a first voice command representing a request to return content of a webpage without incurring cellular data usage, the webpage having one or more topics, wherein the cellular voice connection is initiated as an outgoing call from the mobile device in response to the mobile device meeting a cellular data usage limit;

receiving, from the mobile device, a uniform resource locator (URL) for the webpage, wherein the URL is extracted from a web browser on the mobile device and utilized by an application program on the mobile device;

identifying, based on a natural processing language (NLP) evaluation of the webpage, one or more candidates from the topics of the webpage that match the content specified in the request by;

classifying, via execution of a pipeline in a deep question-answering (QA) system, the first voice command as a question to the deep QA system,

identifying, via the execution of the pipeline, a context for the question,

retrieving one or more features from an electronic storage, based on the context, wherein the features are predictive in determining a response to the question, and

determining the one or more candidates based on the one or more features;

determining a confidence score for each candidate based on the NLP evaluation;

prompting an audio indication requesting a selection from at least two of the candidates, wherein the confidence score of the at least two of the candidates is equal;

receiving a second voice command representing the selection from the at least two of the candidates; and

returning, via the cellular voice connection, an audio indication of the content corresponding to the selected candidate.

2. The method of claim 1, wherein the request specifies a uniform resource locator (URL) for the webpage.

3. The method of claim 2, further comprising, prior to receiving the first voice command:

accessing the webpage via the specified URL;

extracting unstructured text from the webpage; and

identifying the topics of the webpage.

4. The method of claim 1, further comprising, prior to identifying the one or more candidates:

converting the second voice command to text input.

5. The method of claim 4, wherein identifying one or more features of the webpage that match the content specified in the request comprises:

performing one or more NLP techniques to the converted text input to identify a specified command and a target indicating a location within the webpage; and

identifying the content based on the target.

6. The method of claim 1, wherein the audio indication of the content is a prompt requesting a selection of one or more locations to navigate within the webpage.

7. The method of claim 1, further comprising:

tracking cellular data usage by the mobile device;

determining that the mobile device is approaching the cellular data usage limit;

generating a notification to the user, in response to the determination that the mobile device is approaching the cellular data usage limit;

generating a prompt to the user, in response to a determination that the mobile device has met the cellular data usage limit; and

receiving an indication in response to the prompt to the user, wherein the cellular voice connection is initiated in response to receiving the indication.