System and method for remote speech recognition

• US 8,666,741 B2
• Filed: 08/04/2008
• Issued: 03/04/2014
• Est. Priority Date: 11/13/2002
• Status: Active Grant

• Current Assignee: Chiharu Technologies LLC
• Original Assignee: Chiharu Technologies LLC
• Inventors: Reynolds, Douglas F., Knott, Benjamin Anthony, Bushey, Robert Randal
• Primary Examiner(s): Abebe, Daniel D
• Application Number:
  US12/185,670
• Publication Number:
  US 20080294435A1
• Time in Patent Office: 2,038 Days
• Field of Search: 704/235, 704/275, 379/88.01
• US Class Current: 704/235
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

A system and method for remote speech recognition includes one or more customer premise equipment, a speech engine, and a communication engine. The customer premise equipment interfaces with a host from which the customer premise equipment is remotely located. The speech engine, remotely located from the host, recognizes a plurality of speech spoken by a user of the customer premise equipment and translates the speech into the language of the host. The speech engine further converts the recognized speech into one or more text data packets where the text data packets include the recognized speech as data instead of voice. The communication engine encrypts the text data packets and transmits the text data packets to the host. Transmitting data instead of voice to the host reduces the computational demands on the host. Additionally, the communication engine receives a plurality of information from the host.

Claims

1. A system for remote speech recognition, the system comprising:

customer premise equipment remote from a host configured to support a host language, the customer premise equipment configured to interface with the host, and wherein the customer premise equipment includes;

a speech recognition engine customizable based on a characteristic of a user of the customer premise equipment and configured to;

recognize and verify an identity of the user;

recognize user speech from the user;

convert the user speech into text data packets, wherein the text data packets include text corresponding to the user speech; and

a communication engine associated with the speech recognition engine wherein the communication engine is configured to;

receive the text data packets from the speech recognition engine;

generate encrypted packets by encrypting the text data packets; and

transmit the encrypted packets over a communication network to the host.

2. A method for remote speech recognition, the method comprising:

recognizing, by a processing device located at a customer premise equipment, user speech from a user of the customer premise equipment;

converting, by the processing device, the user speech into text data packets at the customer premise equipment, wherein the text data packets include text corresponding to the user speech;

transmitting, by the processing device, the text data packets over a communication network to a host, wherein the customer premise equipment is remotely located from the host; and

receiving, by the processing device, information from the host based on the user speech in response to transmitting the text data packets.

3. The method of claim 2, wherein receiving the information from the host includes downloading a menu hierarchy from the host.

4. The method of claim 2, wherein receiving the information from the host includes downloading a speech recognition aide from the host.

5. The method of claim 2, wherein recognizing the user speech comprises identifying, by the processing device, a user language from a plurality of spoken languages based, at least in part, on the user speech, wherein the method further comprises translating, by the processing device, the user speech from the identified user language into a host language supported by the host, and wherein the text included in the text data packets is written in the host language.

6. The method of claim 2, further comprising:

customizing, by the processing device, the customer premise equipment based at least in part on a characteristic of the user.

7. The method of claim 6, wherein customizing the customer premise equipment includes monitoring user interaction with the host and the customer premise equipment to obtain the characteristic of the user.

8. The method of claim 6, wherein customizing the customer premise equipment includes initializing the customer premise equipment based on the characteristic of the user.

9. The method of claim 2, further comprising:

recognizing, by the processing device, an identity of the customer premise equipment user; and

verifying, by the processing device, the identity of the customer premise equipment user.

10. The method of claim 9, further comprising:

transmitting, by the processing device, user information to the host after the verifying of the identity of the customer premise equipment user.

11. The method of claim 9, further comprising:

utilizing, by the processing device, biometrics to recognize and verify the identity of the customer premise equipment user.

12. The method of claim 9, wherein verifying the identity of the customer premise equipment user includes:

receiving an indicated password from the customer premise equipment user; and

comparing the indicated password with a previously provided password.

13. The method of claim 2, further comprising anticipating, by the processing device, a user request based on an evaluation of account information associated with the user.

14. A method comprising:

recognizing, by a processing device, user speech from a user of a customer premise equipment;

converting, by the processing device, the user speech into text data packets at the customer premise equipment, wherein the text data packets include text corresponding to the user speech;

encrypting, by the processing device, the text data packets;

transmitting the encrypted text data packets over a communication network to a host; and

receiving information from the host based on the user speech.

15. The method of claim 14, further comprising:

routing a call from the customer premise equipment based on the received information.

16. The method of claim 14, wherein recognizing the user speech comprises identifying a user language from a plurality of spoken languages based, at least in part, on the user speech, and wherein converting the user speech comprises translating the user speech from the identified user language into a host language supported by the host, wherein the text included in the text data packets is written in the host language.

17. The method of claim 14, wherein transmitting the encrypted text data packets to the host and receiving the information from the host includes transmitting and receiving in compliance with a voice over Internet protocol.

18. A system for remote speech recognition, the system comprising;

customer premise equipment remote from a host and configured to interface with the host;

a speech recognition engine remotely located from the host and located at the customer premise equipment, wherein the speech recognition engine is configured to;

recognize user speech from a user of the customer premise equipment; and

convert the user speech into text data packets, wherein the text data packets indicate text corresponding to the user speech; and

a communication engine associated with the speech recognition engine and located at the customer premise equipment, wherein the communication engine is configured to;

transmit the text data packets to the host; and

receive information from the host in response to transmitting the text data packets.

19. The system of claim 18, wherein the host comprises a computer located in a call center.

20. The system of claim 18, wherein the customer premise equipment comprises a telephone.

21. The system of claim 18, wherein the communication engine is further configured to encrypt the text data packets, and wherein the text data packets are transmitted to the host as encrypted text data packets.

22. The system of claim 18, wherein the communication engine is further configured to:

initiate an interactive voice response (IVR) communication session with the host; and

receive a query from the host as part of the IVR communication session, wherein the text data packets are transmitted to the host in response to receiving the query.

23. The system of claim 18, wherein the speech recognition engine is further configured to:

identify a user language from a plurality of spoken languages based, at least in part, on the user speech; and

translate the user speech from the identified user language into a host language supported by the host, wherein the text included in the text data packets is written in the host language.

24. The system of claim 18, wherein the speech recognition engine is further configured to monitor a plurality of conversations to obtain one or more user characteristics of the user, and wherein the user speech is converted into text data packets based, at least in part, on the one or more user characteristics.

25. The system of claim 24, wherein the one or more user characteristics are based on a speech pattern of the user.