Off site voice enrollment on a transcription device for speech recognition
AssigneeInternational Business Machines Corporation
InventorsBrian S. Brooks, Waltraud Brunner, Carmi Gazit, Arthur Keller, Antonio R. Lee, Thomas Netousek, Kerry A. Ortega
Life of this patent
Estimating the term…
Abstract
A method and system for use with a computer recognition system to enroll a user. The method involves a series of steps. The invention provides a user with an enrollment script. The invention then receives a recording made with a transcription device of a dictation session in which the user has dictated at least a portion of the enrollment script. Additionally, the invention can enroll the user in the speech recognition system by decoding the recording and training the speech recognition system.
Claims15
In a computer speech recognition system, a method for enrollment of a user comprising:
- providing a user with an enrollment script;
- uploading a recording of a dictation session in which said user has dictated at least a portion of said enrollment script, wherein said recording is made with, and uploaded from a portable transcription device;
- enrolling said user in said speech recognition system by decoding said recording and training said speech recognition system to construct at least one location specific and transcription device specific personal voice model;
- and,
- enabling a user activatable icon for initiating said training process using said recording, if said recording contains a minimum predetermined percentage of words.
In a computer speech recognition system, a system for enrollment of a user comprising:
- means for providing a user with an enrollment script;
- means for uploading a recording of a dictation session in which said user has dictated at least a portion of said enrollment script, wherein said recording is made with, and uploaded from a portable transcription device;
- means for enrolling said user in said speech recognition system by decoding said recording and training said speech recognition system to construct at least one location specific and transcription device specific personal voice model;
- and,
- means for enabling a user activatable icon for initiating said training process using said recording, if said recording contains a minimum predetermined percentage of words.
A machine readable storage, having stored thereon a computer program having a plurality of code sections executable by a machine for causing the machine to perform the steps of:
- providing a user with an enrollment script;
- uploading a recording of a dictation session in which said user has dictated at least a portion of said enrollment script, wherein said recording is made with, and uploaded from a portable transcription device;
- enrolling said user in said speech recognition system by decoding said recording and training said speech recognition system to construct at least one location specific and transcription device specific personal voice model;
- and,
- enabling a user activatable icon for initiating said training process using said recording, if said recording contains a minimum predetermined percentage of words.
Dependencies are inferred from the claim text.