System for transferring personalize matter from one computer to another

• US 7,689,416 B1
• Filed: 01/23/2004
• Issued: 03/30/2010
• Est. Priority Date: 09/29/1999
• Status: Expired due to Fees

• Current Assignee: SynKloud Technologies, LLC (SY Ventures)
• Original Assignee: Darrell A. Poirier
• Inventors: Poirier, Darrell A.
• Primary Examiner(s): Chawan, Vijay B
• Application Number:
  US10/763,966
• Time in Patent Office: 2,258 Days
• Field of Search: 704/235, 704/260, 704270-275, 704/9, 704/231, 704/246, 704/251, 704/257
• US Class Current: 704/235
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

This invention combines methodologies that enhance voice recognition dictation. It describes features for moving speaker voice files eliminating redundant training of speech recognition dictation applications. It defines how to create synthetic voice models reducing speaker dependency. It combines accuracy and performance into a single measure called RAP Rate. Moreover, the invention describes enhancing voice recognition applications and systems by measure/adjusting hardware and software features for optimal voice recognition dictation incorporating methodical processes based on RAP Rate. Using these approaches and tools the invention includes a method for constructing a handheld transcriber that immediately translates audio speech into text with real-time display. The invention describes a method for applying RAP Rate and synthetic voice models to applications like voice mail to text. With the ability to move and translate voice models the invention describes new services that could be provided for a fee.

Claims

1. A computer software application that functions upon a first computer that uses a speech recognition computer program, into which program a first user is enrolled and which program has produced the first user'"'"'s user voice model files and the first user'"'"'s enrollment entries, thereby allowing the first user access to the program and increasing the accuracy of the said program, the said application comprising:

a. a recording means adapted to record the first user'"'"'s voice model files and enrollment entries, stored on the memory of the first computer, and store the first user'"'"'s voice model files and enrollment entries in recoverable form as the first user'"'"'s stored enrollment entries,

b. a transfer means adapted to move the first user'"'"'s user voice model files and the first user'"'"'s stored enrollment entries to a computer file storage device, and

c. an enrollment means adapted to enroll the first user on another computer other than the first computer, said another computer having a speech recognition computer program installed on it, using the first user'"'"'s stored enrollment entries and the first user'"'"'s user voice model files from the computer file storage device, to thereby allow the first user to have access to and increase the accuracy of the speech recognition computer program installed on the said another computer.

2. A software application as recited in claim 1 that includes an identification means that can identify that speech recognition computer program installed on a computer.

3. A software application as recited in claim 1 that can identify users that are enrolled in the speech recognition computer program installed on a computer.

4. A software application as recited in claim 1 that can copy speech recognition training files, system registry parameters, vocabularies, and macros, and package them as a “

Voice Model”

separately from the speech recognition computer program.

5. A software application as recited in claim 1 that can automatically or manually move a Voice Model to another system or physical location.

6. A software application as recited in claim 1 that can move, install, and enroll a Voice Model including all the specific user parameters, files, and vocabularies from the said first computer to the said another computer.

7. A software application as recited in claim 1 that obviates the need for training the speech recognition program on the said another computer.

8. A computer software application that functions upon a first computer that uses a speech recognition computer program, into which program a first user is enrolled and which program has produced a first user'"'"'s user voice model files and a first user'"'"'s enrollment entries, said application comprising:

a. a recording means adapted to record the first user'"'"'s voice model files and enrollment entries, stored on the first computer, and store the first user'"'"'s enrollment entries in recoverable form as the first user'"'"'s voice model files and enrollment entries,

b. a transfer means adapted to move the first user'"'"'s user voice model files and the first user'"'"'s stored enrollment entries to a second computer having a speech recognition computer program installed on it, and

c. an enrollment means adapted to enroll the first user on said second computer, using the first user'"'"'s stored enrollment entries and the first user'"'"'s user voice model files transferred from the first computer.

9. A software application as recited in claim 8 that includes an identification means that can identify that speech computer recognition program installed on a computer.

10. A software application as recited in claim 8 that can identify users that are enrolled in the speech recognition computer program installed on a computer.

11. A software application as recited in claim 8 that can copy speech recognition training files, system registry parameters, vocabularies, and macros, and package them as a “

Voice Model”

separately from the speech recognition program.

12. A software application as recited in claim 8 that can automatically

or manually move a Voice Model to another system or physical location.

13. A software application as recited in claim 8 that can move, install,

and enroll a Voice Model including all the specific user parameters, files, and vocabularies from the said first computer to the said another computer.

14. A software application as recited in claim 8 that obviates the need for training the speech recognition computer program on the said another computer.

15. A computer software application that functions upon a first computer that uses a speech recognition computer program, into which program a first user is enrolled and which program has produced a first user'"'"'s user voice model files and a first user'"'"'s enrollment entries, said application comprising:

a. a recording means adapted to record the first user'"'"'s voice models and enrollment entries, stored on the first computer, and store the first user'"'"'s enrollment entries in recoverable form as the first user'"'"'s stored voice model and enrollment entries,

b. a transfer means adapted to move the first user'"'"'s user voice model files and the first user'"'"'s stored enrollment entries to a computer file storage device, and to move the first user'"'"'s user voice model files and the first user'"'"'s stored enrollment entries from the computer file storage device to another computer other than the first computer, said another computer having a speech recognition computer program installed on it, and

c. an enrollment means adapted to enroll the first user on the said another computer, using the first user'"'"'s stored enrollment entries and the first user'"'"'s user voice model files from the computer file storage device.

16. A software application as recited in claim 15 that includes an identification means that can identify that speech recognition computer program installed on a computer.

17. A software application as recited in claim 15 that can identify users that are enrolled in the speech recognition computer program installed on a computer.

18. A software application as recited in claim 15 that can copy speech recognition training files, system registry parameters, vocabularies, and macros, and package them as a “

Voice Model”

separately from the speech recognition computer program.

19. A software application as recited in claim 15 that can automatically or manually move a Voice Model to another system or physical location.

20. A software application as recited in claim 15 that can move, install, and enroll a Voice Model including all the specific user parameters, files, and vocabularies from the said first computer to the said another computer.

21. A software application as recited in claim 15 that obviates the need for training the speech recognition computer program on the said another computer.

22. A method for training a second speech recognition computer, so that the second speech recognition computer is more effective at speech recognition than it was prior to the training, comprising the steps of:

a) training a first speech recognition computer by causing the said first speech recognition computer to form first voice model files that contain the training result, and thereby to cause the first speech recognition computer to be more effective at speech recognition,

b) making the said first voice model files accessible to a second speech recognition computer, that is adapted to use the said first voice model files to enhance the effectiveness of the said second speech recognition computer at speech recognition, and

c) causing the second speech recognition computer to use the said first voice model files to enhance the effectiveness of said second speech recognition computer at speech recognition.

23. A speech recognition computer system, comprising:

a) a first computer including a memory, said memory containing a first installation of a speech recognition program, and said memory also containing first voice model files that are adapted to be used by the said speech recognition program to increase the accuracy of the said speech recognition program when the said first computer is used by a user to convert the speech of the user into text, said speech recognition program also containing a training module adapted to convert generic voice model files to the first voice model files by monitoring the conversion of the speech of the user into text,

b) a second computer including a memory, said memory containing a second installation of the speech recognition program, and

c) a transfer device adapted to copy the first voice model files from the first computer to the second computer in such a way that the first voice model files can be utilized by the second installation of the speech recognition program to increase the accuracy of the second installation of the speech recognition program when the said second installation of the speech recognition program is used by the user.