International Business Machines Corporation

Barbara E Ballard

James Robert Lewis

Kerry Alan Ortega

Arthur Keller

Huifang Wang

Gary Robert Hanson

Ronald E VanBuskirk

Ronald E Vanbuskirk

Simon C Chu

Amado Nassiff

Linda M Boyer

Ronald Van Buskirk

Thomas A Kist

Gayle Colby Steinbugler

Ronald E Van Buskirk

A novel apparatus and method for correcting speech recognized text in a predominantly speech-only environment for use with a device having only a limited or no display device available. The method is preferably implemented by a machine readable storage mechanism having stored thereon a computer program, the method comprising the following steps. First, audio speech input can be received and speech-to-text converted to speech recognized text. Second, a first speech correction command for performing a correction operation on speech recognized text stored in a text buffer can be detected in the speech recognized text. Third, if a speech correction command is not detected in the speech recognized text, the speech recognized text can be added to the text buffer. Fourth, if a speech command is detected in the speech recognized text, the detected correction speech command can be performed on speech recognized text stored in the text buffer.

Speech recognition correction for devices having limited or no display

A method of proofreading and correcting dictated text contained in an electronic document can include selecting proofreading criteria for identifying textual errors contained in the electronic document; playing back each word contained in the electronic document; and, marking as a textual error each played back word in nonconformity with at least one of the proofreading criteria.

Method and system for proofreading and correcting dictated text

A method of proofreading and correcting dictated text contained in an electronic document comprises the steps of: selecting proofreading criteria for identifying textual errors contained in the electronic document; playing back each word contained in the electronic document; and, marking as a textual error each played back word in nonconformity with at least one of the proofreading criteria. The method can further comprise the step of editing each the marked textual error identified in the marking step. In particular, the editing step can include reviewing each the marked textual error identified in the marking step; accepting user specified changes to each marked textual error reviewed in the reviewing step; and, unmarking each marked textual error corrected by the user in the accepting step. Also, the reviewing step can include highlighting each the word in the electronic document corresponding to the marked textual error marked in the marking step; and, displaying an explanation for each marked textual error in a user interface. Moreover, the reviewing step can further include suggesting a recommended change to the marked textual error; displaying the recommended change in the user interface; and, accepting a user specified preference to substitute the recommended change for the marked textual error.

In a computer speech dictation system, a method for automatically managing a plurality of acoustic models. The method is intended to ensure that only high reliability acoustic models which accurately reflect word pronunciations for a given user are retained. The method is accomplished by assigning a base quality metric value for each of the acoustic models maintained by a speech recognition application. The quality metric is incremented or decremented upon the occurrence of certain events relevant to the reliablity of the acoustic model. Acoustic model are discarded when the quality metric value falls below a threshold value.

Method and apparatus for intelligently managing multiple pronunciations for a speech recognition vocabulary

A method and system are disclosed for managing a plurality of data storage systems. A plurality of graphically designatable data storage systems are displayed within a computer system display. A subset of the plurality of graphically designatable data storage systems are selected to be utilized in a data storage array. A data distribution mode for the data array is designated. An iconic representation of selected data storage systems of the data storage array is automatically graphically displayed in response to the selected subset of data storage systems and the designated data distribution mode such that complete data storage arrays may be graphically illustrated for management purposes.

User interface for configuring and managing a multi-drive data storage system

A method and system for improving the speech command recognition accuracy of a computer speech recognition system uses event-based constraints to recognize a spoken command. The constraints are system states and events, which include system activities, active applications, prior commands and an event queue. The method and system is performed by monitoring events and states of the computer system and receiving a processed command corresponding to the spoken command. The processed command is statistically analyzed in light of the system events and states as well as according to an acoustic model. The system then identifies a recognized command corresponding to the spoken command.

Method and apparatus for improving speech command recognition accuracy using event-based constraints

A method and system for responding to randomly occurring noise in a voice recognition application program. The system receives an audio signal representative of sound in an audio environment and processes the audio signal to identify certain non-speech sounds. A pre-defined action is performed in response to the non-speech sound which has been identified. The pre-defined action is selected from the group consisting of disabling a microphone source of the audio signal, suspending further processing of the audio signal by the speech recognition system, executing a user-defined macro, and ignoring the sound. The system may perform additional steps including recording a sound which is to be identified as a non-speech sound and assigning one of the pre-defined actions to be performed in response when the non-speech sound has been identified.

Noise recognizer for speech recognition systems

In a computer system having a word processing application adapted for speech recognition, a method and system to propagate corrections in dictated text. in response to a first user command, a first error word is selected to be corrected from a word processing file containing dictated text. The error word is replaced by a correct word selected by a user. The system subsequently searches the dictated text contained in the document fie in response to another user command to locate a second or subsequent error word that is a close acoustic match to the first error word. The second or subsequent error word may then be replaced using a word selected from an alternate word list.

Method and apparatus for propagating corrections in speech recognition software

In a computer speech dictation system, a method and system for automatically determining whether to update a language model based upon user amendments to text in a dictated word processing document. The method includes the steps of monitoring a user's text-amending input to an application program for amending the dictated text and subsequently performing processing responsive to the monitoring step to determine whether the text-amending input corrects a misrecognized word in the dictated text. If the test amending input corrects a misrecognized word, the system updates a language model based upon the text which has been amended.

Method and system for automatically determining whether to update a language model based upon user amendments to dictated text

A method for correcting misrecognition errors comprises the steps of: dictating to a speech application; marking misrecognized words during the dictating step; and, after the dictating and marking steps, displaying and correcting the marked misrecognized words, whereby the correcting of the misrecognized words is deferred until after the dictating step is concluded and the dictating step is not significantly interrupted. The displaying and correcting step can be implemented by invoking a correction tool of the speech application, whereby the correcting of the misrecognized words trains the speech application.

Marking and deferring correction of misrecognition errors

A method for managing a What Can I Say (WCIS) function in an application having a plurality of commands which can be voice activated comprises the steps of: storing a set of substantially all voice activatable commands associated with the application; identifying those of the commands in the set which are displayable by the application; and, in response to a user input, displaying in a graphical user interface (GUI) a subset of the voice activatable commands which are not displayable by the application. Moreover, the method includes displaying in the GUI, in response to a user input, a list of the stored set of substantially all voice activatable commands associated with the application; displaying in the GUI a pull down menu identifying different categories by which the commands can be viewed in the list; and, displaying the GUI with a pull down menu identifying commands that can be performed against a voice command.

Managing voice commands in speech applications

A method for correcting frequently misrecognized words and commands in a speech application. According to the method, when a need for correcting a frequently misrecognized word/command spoken by a user is detected, a recording is made of the misrecognized word/command in isolation. Subsequently, an in-isolation base form for the misrecognized word/command is established from the in-isolation recording. The in-isolation base form is then saved and the misrecognized word/command in recorded in context. Next, an in-context base form is established for the misrecognized word/command from the context recording and a comparison is made between the in-isolation and in-context base forms. The in-context base form is saved only if the in-isolation and in-context base forms are markedly different from one another. A sentence is displayed using the frequently misrecognized word/command in context and the user is prompted to speak the sentence. The sentence is then recognized using the speech application and the user is prompted to confirm whether or not the frequently misrecognized word/command was properly recognized. The method is terminated if the frequently misrecognized word/command was properly recognized.

Growing community of inventors

Kansas City, MO, United States of America

Barbara E Ballard