Speech recognition that adjusts automatically to input devices

• US 6,505,161 B1
• Filed: 05/01/2000
• Issued: 01/07/2003
• Est. Priority Date: 05/01/2000
• Status: Expired due to Fees

• Current Assignee: Sprint Spectrum LP (Deutsche Telekom AG)
• Original Assignee: Sprint Communications Company LP (Deutsche Telekom AG)
• Inventors: Brems, Douglas J.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): NOLAN, DANIEL A
• Application Number:
  US09/564,393
• Time in Patent Office: 981 Days
• Field of Search: 704/231, 704/251, 704/255, 704/270-275, 704/270.1, 704/258, 704/246
• US Class Current: 704/270
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering

Abstract

A system and method for customizing the operating characteristics of a speech recognition system with characteristics of a user device. The user device transmits information representing the operating characteristics of itself to the speech recognition system. The speech recognition system determines the speech characteristics of the user device from this information. The speech recognition system obtains the speech characteristics relating to the device from a database and configures the speech recognition system with these characteristics.

Claims

1. A method for customizing configurable parameters of a speech recognition system with operating characteristics associated with a user device comprising the steps of:

supplying a user device, said user device having a device type;

supplying a speech recognition system, said speech recognition system having configurable parameters;

transmitting information representing said device type of said user device from said user device to said speech recognition system;

supplying a database and storing in said database said operating characteristics associated with said user device of said device type;

obtaining said operating characteristics of said user device from said database based upon said information representing said device type; and

configuring said configurable parameters of said speech recognition system with said operating characteristics.

2. The method of claim 1 comprising the further steps of:

receiving a voice message from said user device; and

recognizing the contents of said voice message using said operating characteristics.

3. The method of claim 2 comprising the further step of:

activating a service in response to recognizing the contents of said voice message.

4. The method of claim 3 wherein said service is a voice mail system.

5. The method of claim 3 wherein said contents of said voice message requests a call to be outdialed and whereby the method further comprises the steps of determining connection information for said call and outdialing said call based upon said connection information.

6. A method for configuring a speech recognition unit comprising the steps of:

supplying a user device having operating characteristics;

supplying a speech recognition unit having configurable parameters;

sending a presence message from said user device to said speech recognition unit, wherein said presence message indicates the need for service from said user device;

responding to said presence message with an acknowledgement message;

in response to said acknowledgement message, determining a code which describes said operating characteristics of said user device;

sending said code to said speech recognition unit in a device-type message;

receiving said device-type message and extracting said code from said device-type message;

in response to said step of receiving said device-type message, determining a speech characteristic model corresponding to said code;

sending said speech characteristic model to said speech recognition unit;

configuring said configurable parameters of said speech recognition unit with said speech characteristic model;

sending a voice message whereby said voice message contains instructions to dial a number;

processing said voice message and determining said number; and

outdialing a call to said number.

7. A speech recognition system comprising:

a plurality of user devices, each of said user devices having an associated device type, said user devices comprising cellular and landline phones;

a database, said database storing speech characteristics based upon said type of said device;

a communication network coupled to said user devices and receiving messages from said user devices, said messages specifying said device type, said plurality of user devices communicatively coupled to said network;

speech recognition means coupled to said database and said communication network for receiving messages from said network and having updateable parameters;

speech adaptation means coupled to said speech recognition means comprising a speech recognition unit for automatically updating said parameters of said speech recognition means with speech models and parameters corresponding to the type of selected ones of said plurality of user devices; and

service means for performing network services, said service means being activated by said speech adaptation means, said service means including means for out-dialing phone calls and accessing a voicemail system.

8. A speech recognition system comprising:

a user device, said user device having an associated type;

a database, said database storing speech characteristics based upon said type of said user device;

a communication network coupled to said user device and receiving messages from said user device, said user device communicatively coupled to said communication network;

speech recognition means coupled to said database and said communication network for receiving messages from said network, said speech recognition means having configurable parameters; and

speech adaptation means for automatically updating said configurable parameters of said speech recognition means with said speech characteristics from said database, said speech characteristics corresponding to said type of said user device.

9. The system of claim 8 further comprising service means for performing network services, said service means being activated by said speech adaptation means.

10. The system of claim 9 wherein said service means places a telephone call.

11. The system of claim 9 wherein said service means comprises a voice mail system.

12. The system of claim 8 wherein said speech characteristics comprise a plurality of speech models and feature parameters.

13. The system of claim 8 wherein said user device is a selected one of a landline handset, landline speaker phone, cellular handset, or cellular speaker phone.

14. The system of claim 8 wherein said user device is a telematics unit.

15. The system of claim 8 wherein said speech adaptation means includes a supervisor, said supervisor receiving messages from said user device, said messages specifying the type of said user device.

16. A system for updating parameters of a speech recognition unit comprising:

a user device, said user device having an associated type;

a database, said database storing speech characteristics based upon said type of said user device;

a communication network coupled to said user device and receiving messages from said user device, said user device communicatively coupled to said communication network;

a speech recognition unit, said speech recognition unit receiving messages from said communication network, said speech recognition unit coupled to said communication network and said database, said speech recognition unit having configurable parameters, said speech recognition unit automatically updating said configurable parameters with said speech characteristics from said database, said characteristics corresponding to said type of said user device.

17. The system of claim 16 further comprising a services module, said services module performing network services.

18. The system of claim 17 wherein said services module places a telephone call.

19. The system of claim 17 wherein said services module comprises a voice mail system.

20. The system of claim 16 wherein said speech characteristics comprise a plurality of speech models and feature parameters.

21. The system of claim 16 wherein said user device is a selected one of a landline handset, landline speaker phone, cellular handset, or cellular speaker phone.

22. The system of claim 16 wherein said user device is a telematics unit.

23. The system of claim 16 wherein said communication network further comprises a supervisor, said supervisor receiving messages from said user device, said messages specifying the type of said user device.

24. A method for customizing the operating characteristics of a speech recognition system with characteristics of a user device comprising the steps of:

supplying a user device having a device type;

supplying a speech recognition unit having updateable parameters;

transmitting a message from said user device requesting service;

responding to said message with an answer wherein said answer grants service to said user device;

transmitting information representing said device type of said user device from said user device to said speech recognition system;

determining the speech characteristics of said user device from said information;

obtaining the speech models and parameters relating to said user device from a database;

and

updating the parameters of said speech recognition unit with said speech characteristics.

25. The method of claim 24 wherein said user device is a selected one of a landline handset, landline speaker phone, cellular handset, or cellular speaker phone.

26. The method of claim 24 comprising the further steps of:

receiving a voice message from said user device; and

recognizing the contents of said voice message.

27. The method of claim 26 comprising the further step of:

activating services in response to recognizing the contents of said voice message.

28. The method of claim 27 wherein said contents of said voice message requests a call to be outdialed and whereby the method further comprises the steps of determining connection information for said call and outdialing a call based upon said connection information.

29. The method of claim 27 wherein said services include a voice mail system.