Apparatus and methods for implementing voice enabling applications in a converged voice and data network environment

• US 7,805,310 B2
• Filed: 02/26/2002
• Issued: 09/28/2010
• Est. Priority Date: 02/26/2001
• Status: Active Grant

• Current Assignee: Elizabeth A. Rohwer
• Original Assignee: Elizabeth A. Rohwer
• Inventors: Rohwer, Elizabeth A.
• Primary Examiner(s): Armstrong; Angela A
• Application Number:
  US10/469,110
• Publication Number:
  US 20040107108A1
• Time in Patent Office: 3,136 Days
• Field of Search: 704/270.1, 704/270, 704/273, 704/275, 704/231, 704/246, 704/247, 704/250, 713/186
• US Class Current: 704/270.1
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: H04M 2201/41 using speaker recognition speaker recognition per se G10L17/00
• CPC Class Codes: H04M 2201/42 Graphical user interfaces
• CPC Class Codes: H04M 7/006 Networks other than PSTN/ISDN providing telephone service, e.g. Voice over Internet Protocol (VoIP) , including next generation networks with a packet-switched transport layer H04L65/00 takes precedence; aspects not specific to the type of network H04M3/00; special services in those networks H04M3/42

Abstract

Human speech is transported through a voice and data converged Internet network to recognize its content, verify the identity of the speaker, or to verify the content of a spoken phrase by utilizing the Internet protocol to transmit voice packets. The voice data (4) entered is processed and transmitted in the same way as Internet data packets over converged voice and data IP networks. A voice-enabled application sends a message (5), which is decoded by the speech API (2) and the appropriate control and synchronization information is issued (7) to the data preparation module (9) and to the speech engine (3). Standard voice over IP includes a speech compression algorithm and the use of RTP (Real Time Protocol), enabling additional processing of the human voice anywhere in the network to perform speaker verification, with or without the knowledge of the speaker.

Claims

1. A method of implementing voice-enabled applications in a single converged voice and data network IP environment that supports and implements VoIP connections, protocols and traffic, the method comprising:

a. entering human voice data as data packets into a converged VoIP packet-switched IP based voice and data network;

b. converting the voice data into a digital format for VoIP transport and speech processing at a later time;

c. providing a control mechanism to interface at least one packet-switched network voice enabled application with at least one speech processing function using a VoIP speech application interface;

d. performing at least one speech processing biometrics function inside the VoIP network using the voice of the end user, wherein the voice-enabled application performs non-intrusive voice ID biometrics during which the user enters his voice into the network for the purpose of performing a VoIP telephone conversation and a sample from the voice of the telephone conversation is used to verify the user in a manner transparent to the user; and

e. taking an action with respect to the end user depending on the result from the speech processing of the voice data, said action being taken from the group consisting of allowing the end user to enter into a secured transaction, awarding the end user with a prize, penalizing the end user by disallowing entry, and providing feedback to the end user to communicate the result.

2. A method of implementing voice-enabled applications in a single converged voice and data network IP environment that supports and implements VoIP connections, protocols and traffic, the method comprising:

a. entering human voice data as data packets into a converged VoIP packet-switched IP based voice and data network;

b. converting the voice data into a digital format for VoIP transport and speech processing at a later time;

c. providing a control mechanism to interface at least one packet-switched network voice enabled application with at least one speech processing function using a VoIP speech application interface;

d. performing at least one speech processing biometrics function inside the VoIP network using the voice of the end user;

wherein the voice-enabled application residing in a packet-switched IP based network for VoIP traffic performs spoken text verification without an intrusive response, where the computer screen prompts the user to speak an advertising phrase and the voice enters the network through the computer without passing through a PSTN network; and

e. taking an action with respect to the end user depending on the result from the speech processing of the voice data, said action being taken from the group consisting of allowing the end user to enter into a secured transaction, awarding the end user with a prize, penalizing the end user by disallowing entry, and providing feedback to the end user to communicate the result.

3. A method of implementing voice-enabled applications in a single converged voice and data network IP environment that supports and implements VoIP connections, protocols and traffic, the method comprising:

a. entering human voice data as data packets into a converged VoIP packet-switched IP based voice and data network;

b. converting the voice data into a digital format for VoIP transport and speech processing at a later time;

c. providing a control mechanism to interface at least one packet-switched network voice enabled application with at least one speech processing function using a VoIP speech application interface;

d. performing at least one speech processing biometrics function inside the VoIP network using the voice of the end user, wherein the voice-enabled application residing in a packet-switched IP based network for VoIP traffic performs spoken text verification where the computer screen prompts the user to speak an advertising phrase while simultaneously verifying the user'"'"'s identity through non-intrusive voice ID biometrics, and the voice enters the network through the computer without passing through a PSTN network; and

e. taking an action with respect to the end user depending on the result from the speech processing of the voice data, said action being taken from the group consisting of allowing the end user to enter into a secured transaction, awarding the end user with a prize, penalizing the end user by disallowing entry, and providing feedback to the end user to communicate the result.

4. A method of implementing voice-enabled applications in a single converged voice and data network IP environment that supports and implements VoIP connections, protocols and traffic, the method comprising:

a. entering human voice data as data packets into a converged VoIP packet-switched IP based voice and data network;

b. converting the voice data into a digital format for VoIP transport and speech processing at a later time;

c. providing a control mechanism to interface at least one packet-switched network voice enabled application with at least one speech processing function using a VoIP speech application interface;

d. performing at least one speech processing biometrics function inside the VoIP network using the voice of the end user, wherein the voice-enabled application residing in a packet-switched IP based network for VoIP traffic performs non-intrusive voice ID biometrics simultaneously to more than one speaker participating in a conference call and indicates who is currently speaking by highlighting a reference to him on a monitoring WEB site; and

e. taking an action with respect to the end user depending on the result from the speech processing of the voice data, said action being taken from the group consisting of allowing the end user to enter into a secured transaction, awarding the end user with a prize, penalizing the end user by disallowing entry, and providing feedback to the end user to communicate the result.

5. A method of implementing voice-enabled applications in a single converged voice and data network IP environment that supports and implements VoIP connections, protocols and traffic, the method comprising:

a. entering human voice data as data packets into a converged VoIP packet-switched IP based voice and data network;

b. converting the voice data into a digital format for VoIP transport and speech processing at a later time;

c. providing a control mechanism to interface at least one packet-switched network voice enabled application with at least one speech processing function using a VoIP speech application interface;

d. performing at least one speech processing biometrics function inside the VoIP network using the voice of the end user, wherein the voice-enabled application residing in a packet-switched IP based network for VoIP traffic performs non-intrusive voice ID biometrics in order to differentiate the voices of the users of a flat rate long distance VoIP telephone service and allow usage-only by the legitimate subscriber; and

e. taking an action with respect to the end user depending on the result from the speech processing of the voice data, said action being taken from the group consisting of allowing the end user to enter into a secured transaction, awarding the end user with a prize, penalizing the end user by disallowing entry, and providing feedback to the end user to communicate the result.

6. A method of implementing voice-enabled applications in a converged voice and data network IP environment that supports and implements VoIP connections, protocols and traffic, comprising:

a. entering human voice data as data packets from a speaker into a converged VoIP voice and data packet-switched IP based network for later processing and acoustic matching;

b. non-intrusively processing the voice data from an end user biometrically while the end user is speaking into the converged VoIP voice and data packet-switched IP based network;

c. preparing the voice data for VoIP speech processing function with a front-end processing module to separate the pauses in the speech from the voice data and utilizing a voice feature extraction module to ready the speech for a processing algorithm from a speech engine;

d. synchronizing generated control information with a data exchange between a data preparation module and the speech engine through a speech application interface;

e. processing the voice features by the speech engine to perform at least one combination of the speech processing and pattern recognition algorithms implemented by the speech engine;

f. providing feedback to the end user to communicate the result from the speech processing; and

g. taking an action responsive to the result from the processing of the voice.

7. The method of claim 6, wherein the human voice is entered by a data preparation module implemented on a buffering device selected from the group consisting of a computer and a specialized hardware, for buffering the data and passing it directly to the data preparation module without transmission over the conventional telephone network and a speech engine implemented as a part of a centralized server.

8. The method of claim 6, further comprising the implementation of a network speech appliance on the buffering device with the speech application interface communicating with the voice-enabled application on the packet-switched IP based VoIP network using a message-based mechanism.

9. The method of claim 6, further comprising the implementation of the network speech appliance on a wireless device with the speech application interface communicating with the voice-enabled application on the packet-switched IP based VoIP network using the message-based mechanism without the voice passing through the conventional telephone network.

10. The method of claim 6, wherein said step of taking an action responsive to the result from the processing of the voice, further comprising:

h. entering into a dialog with the end user with the aim of acquiring additional voice data from the user, in the case when the first voice data entry does not secure sufficient accuracy from the speech processing algorithms;

i. entering into a dialog with the end user with the aim of acquiring additional voice data from the user in order to perform additional speech processing using a different algorithm from the speech engine for a separate purpose;

j. indicating to the user the result of the VoIP speech processing operation; and

k. repeating the above instant steps if the voice enabled application requires it.

11. The method of claim 6, wherein the step of providing the feedback to the end user is accomplished by using a graphical interface on a screen of a device selected from the group consisting of a computer, a mobile phone, a personal handheld device and a specialized hardware, said graphical interface displaying written messages as prompts and results from the speech processing operation.

12. The method of claim 6, wherein the step of providing the feedback to the end user is accomplished by using audio feedback by playing the dialog messages to the end user over a telephone device without the voice passing through the conventional telephone network.

13. The method of claim 6, wherein the step of providing the feedback to the end user is accomplished by using a combination of a graphical interface from a computer screen with an audio feedback from a VoIP telephone connection with the audio passing through the computer.

14. The method of claim 6, wherein said step of processing the voice features by the speech engine is accomplished by performing a function of training a voiceprint for a new user.

15. The method of claim 14, wherein a combination of the steps can be implemented simultaneously using the same sample of the voice of the user by performing voice ID verification on a prompted text that varies with each prompt and simultaneously performing spoken text verification to ensure the end user is saying the asked-for sequence of words.

16. The method of claim 6, wherein said step of processing the voice features by the speech engine is accomplished by performing a function of voice ID verification of an enrolled user.

17. The method of claim 6, wherein said step of processing the voice features by the speech engine is accomplished by performing a function of speech recognition.

18. The method of claim 6, wherein said step of processing the voice features by the speech engine is accomplished by performing a function of spoken text verification.