Systems and methods for dynamic re-configurable speech recognition

• US 20020046022A1
• Filed: 10/10/2001
• Published: 04/18/2002
• Est. Priority Date: 10/13/2000
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Rose, Richard C., Gajic, Bojana
• Granted Patent: US 7,457,750 B2
• Time in Patent Office: Days
• US Class Current: 704/231
• CPC Class Codes: G10L 15/10 using distance or distortion measures between unknown speech and reference templates
• CPC Class Codes: G10L 15/142 Hidden Markov Models [HMMs]
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence

Abstract

Speech recognition models are dynamically re-configurable based on user information, background information such as background noise and transducer information such as transducer response characteristics to provide users with alternate input modes to keyboard text entry. The techniques of dynamic re-configurable speech recognition provide for deployment of speech recognition on small devices such as mobile phones and personal digital assistants as well environments such as office, home or vehicle while maintaining the accuracy of the speech recognition.

Claims

1. A method of dynamic re-configurable speech recognition comprising the steps of:

determining parameters of a background model of a received voice request;

determining parameters of a transducer model;

determining an adapted speech recognition model for a speech recognition model based on at least one of the background model and the transducer model; and

determining information in the voice request based on the adapted speech recognition model.

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

determining at least one sample period;

determining at least one of a new background model and a new transducer model based on the at least one sample period.

3. The method of claim 2, wherein,

the parameters of the background model are determined based on a first sample period;

and

the parameters of the transducer model are determined based on a second sample period.

4. The method of claim 2, further comprising the steps of:

saving at least one of the parameters of the background model and the parameters of the transducer model;

determining the adapted speech recognition model based on the at least one sample period and at least one of the background model and the transducer model.

5. A system for dynamic re-configurable speech recognition comprising:

a controller;

a background model estimation circuit for determining a background model of a voice request based on estimated background parameters and user information;

a transducer model estimation circuit for determining a transducer model of the voice request based on estimated transducer parameters and user information;

a background model adaptation circuit and a transducer model adaptation circuit for determining an adapted speech recognition model based on a speech recognition model and at least one of the background model and the transducer model.

6. The system of claim 5, wherein, the controller determines at least one sample period and based on the at least one sample period activates at least one of the background model estimation circuit and the transducer model estimation circuit.

7. The system of claim 6, wherein,

the background model is determined based on a first sample period;

and

the transducer model is determined based on a second sample period.

8. The system of claim 6, wherein the controller saves at least one of the background model and the transducer model into storage;

and wherein the adapted speech recognition model is based on the at least one sample period and at least one of the background model and the transducer model.

9. A carrier wave encoded to transmit a control program usable for dynamic re-configurable speech recognition to a device for executing the control program, the control program comprising:

instructions for determining parameters of a background model of a received voice request;

instructions for determining parameters of a transducer model;

instructions for determining an adapted speech recognition model for a speech recognition model based on at least one of the background model and the transducer model; and

instructions for determining information in the voice request based on the adapted speech recognition model.

10. The carrier wave of claim 9, further comprising the steps of:

instructions for determining at least one sample period;

instructions for determining at least one of a new background model and a new transducer model based on the at least one sample period.

11. The carrier wave of claim 10, wherein,

the background model is determined based on the first sample period;

and

the transducer model is determined based on a second sample period.

12. The carrier wave of claim 10, further comprising:

instructions for saving at least one of the background model and the transducer model;

instructions for determining the adapted speech recognition model based on the at least one sample period and at least one of the background model and the transducer model.

13. A computer readable storage medium comprising:

computer readable program code embodied on a computer readable storage medium, said computer readable program code usable to program a computer to perform a method for dynamic re-configurable speech recognition comprising the steps of;

determining parameters of a background model for a received voice request;

determining parameters of a transducer model;

determine an adapted speech recognition model for a speech recognition model based on at least one of the background model and the transducer model; and

determining information in the voice request based on the adapted speech recognition model.

14. A method of dynamic re-configurable speech recognition comprising the steps of:

determining user specific parameters of a background model for a received voice request;

determining user specific parameters of a transducer model;

determine an adapted speech recognition model for a speech recognition model based on at least one of the background model and the transducer model;

determining information in the voice request based on the adapted speech recognition model;

determining at least one sample period;

determining at least one of a new background model and a new transducer model based on the at least one sample period;

wherein, the background model is determined based on a first sample period; and

the transducer model is determined based on a second sample period.