Maintaining input device identity

• US 6,275,805 B1
• Filed: 02/25/1999
• Issued: 08/14/2001
• Est. Priority Date: 02/25/1999
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Buskirk, Ronald Van, Ruback, Harvey, Guasti, Peter, Nassiff, Amado, Fado, Frank
• Primary Examiner(s): Martin-Wallace, Valencia
• Assistant Examiner(s): Harris, Chanda
• Application Number:
  US09/257,673
• Time in Patent Office: 901 Days
• Field of Search: 704/270, 345/133, 345/326, 434/118, 434/379
• US Class Current: 704/270
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A method for maintaining input device identity in a speech application, comprising the steps of: storing a plurality of enrollments, each of the enrollments representing a speech file of training data associated with at least one of a specific audio input device and a specific audio environment for a specific user; generating a graphical user interface (GUI) display screen for prompting and enabling user selection of at least one of an audio input device and an audio environment; and, retrieving one of the enrollments responsive to the user selection, for use in a dictation or transcription session.

Claims

1. A method for maintaining input device identity in a speech application, comprising the steps of:

storing a plurality of enrollments, each of said enrollments representing a speech file of training data associated with at least one of a specific audio input device and a specific audio environment for a specific user;

generating a graphical user interface (GUI) display screen for prompting and enabling user selection of at least one of an audio input device and an audio environment; and

,

retrieving one of said enrollments responsive to said user selection, for use in a dictation or transcription session.

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

after said generating step and prior to said retrieving step, determining whether one of said plurality of enrollments corresponds to said user selection;

in the event not one of said stored enrollments corresponds to said user selection, generating at least one further GUI display screen for prompting and enabling said user to undertake a training session for creating a further enrollment for said user selection; and

,

storing said further enrollment.

3. The method of claim 2, further comprising the step of retrieving said further enrollment for use in said dictation session.

4. The method of claim 1, comprising the step of storing said associations in a system wide data base.

5. The method of claim 1, comprising the step of storing said associations in a Windows registry.

6. The method of claim 1, comprising the steps of:

storing data representing audio configuration settings for sound cards and audio mixers with each of said enrollments; and

,

restoring said audio configuration settings for each retrieved enrollment prior to initiation of said dictation or transcription session.

7. A computer apparatus programmed with a routine set of instructions stored in a fixed medium, for maintaining input device identity in a speech application, comprising:

means for storing a plurality of enrollments, each of said enrollments representing a speech file of training data associated with at least one of a specific audio input device and a specific audio environment for a specific user;

means for generating a graphical user interface (GUI) display screen for prompting and enabling user selection of at least one of an audio input device and an audio environment; and

,

means for retrieving one of said enrollments responsive to said user selection, for use in a dictation or transcription session.

8. The apparatus of claim 7, further comprising:

means, operable after said generating step and prior to said retrieving step, for determining whether one of said plurality of enrollments corresponds to said user selection;

means, operable in the event not one of said stored enrollments corresponds to said user selection, for generating at least one further GUI display screen for prompting and enabling said user to undertake a training session for creating a further enrollment for said user selection; and

,

means for storing said further enrollment.

9. The apparatus of claim 8, further comprising means for retrieving said further enrollment for use in said dictation session.

10. The apparatus of claim 7, wherein said associations are stored in a system wide data base.

11. The apparatus of claim 7, wherein said associations are stored in a Windows registry.

12. The apparatus of claim 7, comprising:

means for storing data representing audio configuration settings for sound cards and audio mixers with each of said enrollments; and

,

means for restoring said audio configuration settings for each retrieved enrollment prior to initiation of said dictation session.