System and methodology for voice activated access to multiple data sources and voice repositories in a single session

• US 6,891,932 B2
• Filed: 12/11/2001
• Issued: 05/10/2005
• Est. Priority Date: 12/11/2001
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Bhargava, Gautam, Kumar, Slva V., Bhasin, Ramneek
• Primary Examiner(s): Tsang, Fan
• Assistant Examiner(s): Gauthier, Gerald
• Application Number:
  US10/015,026
• Publication Number:
  US 20030123618A1
• Time in Patent Office: 1,246 Days
• Field of Search: 379/88.01, 379/88.02, 379/88.03, 379/88.04, 379/88.17, 379/93.24, 455/563, 701/207, 704/1, 704/8, 704/9, 704/275, 704/257, 709/218
• US Class Current: 379/88.02
• CPC Class Codes: H04M 3/4938 comprising a voice browser which renders and interprets, e.g. VoiceXML

Abstract

A system and method and computer program for seamlessly accessing multiple data sources and voice repositories using voice commands in a single phone call session. The system comprises of voice grammars that span various contexts for all data sources and voice repositories, a telephony platform, an automatic speech recognition engine, extractors for extracting information from the data sources and voice repositories and an interpreter for controlling the extractors and telephony platform. It is the co-operation between the voice grammars and the telephony platform, controlled by a VoiceXML interpreter that enables this seamless access to information from the multiple data sources and voice repositories.

Claims

1. A method for providing voice activated seamless access to a plurality of disparate data sources and voice repositories in a single phone call session, the method comprising the steps of:

logging on by a user;

recognizing voice commands issued by the user comprising;

automatically identifying the context of the recognized commands, by analyzing the recognized voice commands;

automatically identifying the data sources and voice repositories required to be accessed, using the identified context; and

automatically identifying the information desired by the user, by analyzing the recognized voice commands, the information pertaining to the identified contexts, without the user being prompted for inputs;

interpreting the voice commands;

extracting the identified information from the identified data sources and voice repositories;

presenting the extracted information to the user;

wherein said logging step further comprises the steps of;

placing a call to a telephony platform by the user;

holding the call on an incoming leg by the telephony platform;

creating an outbound leg by the telephony platform to a server;

directing extractors for a VoiceXML stream for authenticating the user; and

authenticating the user.

2. The method for providing voice activated access to information as recited in claim 1, wherein said directing step is done using Hyper Text Transfer Protocol (HTTP).

3. A method for providing voice activated seamless access to a plurality of disparate data sources and voice repositories in a single phone call session, the method comprising the steps of:

logging on by a user;

recognizing voice commands issued by the user comprising;

automatically identifying the context of the recognized commands, by analyzing the recognized voice commands;

automatically identifying the data sources and voice repositories required to be accessed, using the identified context; and

automatically identifying the information desired by the user, by analyzing the recognized voice commands, the information pertaining to the identified contexts, without the user being prompted for inputs;

interpreting the voice commands;

extracting the identified information from the identified data sources and voice repositories;

presenting the extracted information to the user;

wherein said presenting step further comprises the steps of;

parsing the VoiceXML stream by an interpreter; and

performing an actionable step;

wherein said performing step comprises instructing the telephony platform to place an outbound call, putting the call on hold, and later reconnecting to the call on hold.

4. A computer program product for providing voice activated seamless access to information from a plurality of disparate data sources and voice repositories in a single phone call session, the computer program product embodied on one or more computer readable media and comprising:

a computer readable program code means for logging by a user;

a computer readable program code means for recognizing voice commands issued by the user, the computer program code means comprising;

computer readable program code means for automatically identifying context of the recognized commands by analyzing the recognized voice commands;

computer readable program code means for automatically identifying the data sources and voice repositories required to be accessed, using the identified context; and

computer readable program code means for automatically identifying information desired by the user by analyzing recognized voice commands, the identification being done using the identified contexts without the user being prompted for inputs;

a computer readable program code means for interpreting the voice commands;

a plurality of computer readable program code means for extracting the identified information from the identified data sources and voice repositories;

a computer readable program code means for presenting the extracted information to the user;

wherein said computer readable program code means for logging comprises;

a computer readable program code means for placing a call to by the user;

a computer readable program code means for holding the call on an incoming leg;

a computer readable program code means for creating an outbound leg;

a computer readable program code means for directing extractors for a VoiceXML stream; and

a computer readable program code means for authenticating the user.

5. A computer program product for providing voice activated seamless access to information from a plurality of disparate data sources and voice repositories in a single phone call session, the computer program product embodied on one or more computer readable media and comprising:

a computer readable program code means for logging by a user;

a computer readable program code means for recognizing voice commands issued by the user, the computer program code means comprising;

computer readable program code means for automatically identifying context of the recognized commands by analyzing the recognized voice commands;

computer readable program code means for automatically identifying the data sources and voice repositories required to be accessed, using the identified context; and

computer readable program code means for automatically identifying information desired by the user by analyzing recognized voice commands, the identification being done using the identified contexts without the user being prompted for inputs;

a computer readable program code means for interpreting the voice commands,

a plurality of computer readable program code means for extracting the identified information from the identified data sources and voice repositories;

a computer readable program code means for presenting the extracted information to the user;

wherein said computer readable program code means for presenting the extracted information to the user comprises;

computer readable program code means for parsing the VoiceXML stream by the interpreter; and

computer readable program code means for performing an actionable step;

wherein said computer readable program code means for performing an actionable step comprises computer readable program code means for placing an outbound call for the user.

6. An apparatus for providing voice activated seamless access to a plurality of disparate data sources and voice repositories in a single phone call session, the apparatus comprising:

means for logging on by a user;

means for recognizing voice commands issued by the user comprising;

means for automatically identifying the context of the recognized commands, by analyzing the recognized voice commands;

means for automatically identifying the data sources and voice repositories required to be accessed, using the identified context; and

means for automatically identifying the information desired by the user, by analyzing the recognized voice commands, the information pertaining to the identified contexts, without the user being prompted for inputs;

means for interpreting the voice commands;

means for extracting the identified information from the identified data sources and voice repositories;

means for presenting the extracted information to the user;

wherein said means for logging further comprises;

means for placing a call to a telephony platform by the user;

means for holding the call on an incoming leg by the telephony platform;

means for creating an outbound leg by the telephony platform to a server;

means for directing extractors for a VoiceXML stream for authenticating the user; and

means for authenticating the user.

7. An apparatus for providing voice activated seamless access to a plurality of disparate data sources and voice repositories in a single phone call session, the apparatus comprising:

a processor;

a machine-readable medium including instructions executable by the processor, including;

one or more instructions for logging on by a user;

one or more instructions for recognizing voice commands issued by the user comprising;

one or more instructions for automatically identifying the context of the recognized commands, by analyzing the recognized voice commands;

one or more instructions for automatically identifying the data sources and voice repositories required to be accessed, using the identified context; and

one or more instructions for automatically identifying the information desired by the user, by analyzing the recognized voice commands, the information pertaining to the identified contexts, without the user being prompted for inputs;

one or more instructions for interpreting the voice commands;

one or more instructions for extracting the identified information from the identified data sources and voice repositories;

one or more instructions for presenting the extracted information to the user;

one or more instructions for placing a call to a telephony platform by the user;

one or more instructions for holding the call on an incoming leg by the telephony platform;

one or more instructions for creating an outbound leg by the telephony platform to a server;

one or more instructions for directing extractors for a VoiceXML stream for authenticating the user; and

one or more instructions for authenticating the user.

8. A computer data signal embodied in a carrier wave comprising instructions for providing voice activated seamless access to a plurality of disparate data sources and voice repositories in a single phone call session, including:

one or more instructions for logging on by a user;

one or more instructions for recognizing voice commands issued by the user comprising;

one or more instructions for automatically identifying the context of the recognized commands, by analyzing the recognized voice commands;

one or more instructions for automatically identifying the data sources and voice repositories required to be accessed, using the identified context; and

one or more instructions for automatically identifying the information desired by the user, by analyzing the recognized voice commands, the information pertaining to the identified contexts, without the user being prompted for inputs;

one or more instructions for interpreting the voice commands;

one or more instructions for extracting the identified information from the identified data sources and voice repositories;

one or more instructions for presenting the extracted information to the user;

one or more instructions for placing a call to a telephony platform by the user;

one or more instructions for holding the call on an incoming leg by the telephony platform;

one or more instructions for creating an outbound leg by the telephony platform to a server;

one or more instructions for directing extractors for a VoiceXML stream for authenticating the user; and

one or more instructions for authenticating the user.