Data processing: speech signal processing – linguistics – language – Linguistics
Reexamination Certificate
1999-12-20
2001-07-24
Thomas, Joseph (Department: 2747)
Data processing: speech signal processing, linguistics, language
Linguistics
C704S009000, C340S870030, C400S087000, C434S112000, C434S169000
Reexamination Certificate
active
06266631
ABSTRACT:
FIELD OF THE INVENTION
The invention pertains to the field of Augmentative Communication Technology. The invention addresses the problem of providing a communication device for producing message output reflecting a user's intended message which minimizes the amount of user input required to produce a message.
BACKGROUND OF THE INVENTION
The field of Augmentative Communication Technology (ACT) pertains to the application of technology to the problem of facilitating communication by individuals whose own abilities are not sufficient to allow them to communicate successfully. ACT devices may therefore be understood as “communication prostheses” for individuals with communication disabilities. State of the art ACT devices employ computer and data processing technology to facilitate communication. In general these devices receive input from the user and translate that input into an output message which approximates the message which the user wishes to communicate. Known technology allows messages to be provided in a number of forms, including visual displays and synthesized speech.
A goal in the field of ACT is to increase the rate of communication attainable through ACT devices by reducing the amount of input which is required to produce a desired message. This goal has been pursued to date primarily through the development of user interfaces which employ coding systems to invoke desired messages. In such devices, the ACT device produces output which is associated with a code supplied by the user. For example, a device may associate the code “greeting” with the message “Hello, my name is Jeff.” Thus, when supplied with the input code “greeting”, the device will produce the code's associated message.
A number of coding systems are currently employed by ACT devices. A first type of coding system employs fixed codes which correspond to output messages which may be produced by the system. Various fixed code schemes use truncated words, words without vowels, contractions, or sequences of graphic characters as input codes. For example, in a fixed code system, input consisting of an “apple” icon followed by a “building” icon may represent the word “school”. Coding systems may also use input comprising-combinations of icons and letters.
Fixed code systems generally require the user to input the full code which corresponds to a desired output message. However, some fixed code systems utilize simple predictive algorithms to present the user with a variety of input choices based on the input of a partial code. In one such scheme known as “word prediction”, the user provides a truncated input code and the device displays one or more numbered “candidate” words which may correspond to the full input word intended by the user. Completion of code input is accomplished by the user's selection of the number corresponding to the proper candidate word. Similarly, in icon based coding systems, the input of a first icon of an icon sequence may produce a display of a grid of additional icons which are known by the system to follow the first icon in various input codes.
An alternative coding system employed in some ACT devices is known as variable coding. Variable coding systems are distinguished from fixed coding systems in that codes are not strictly associated with a fixed message. Rather, the message associated with a code may vary depending upon factors such as the frequency or recency of use, or the probable grammatical category of the message. Thus the hierarchy of messages and codes in a variable system is optimized by the system to correspond to the user's history of use.
As an alternative to coding systems, some ACT devices employ databases containing output messages and associated search attributes. The characteristics of a desired output message are indicated by the user in the form of a search string. The device then produces a list of messages having the attributes indicated by the search string, from which the user may select a message for output.
SUMMARY OF THE INVENTION
While present technology enables the devices and methods described above, the performance which present technology provides is limited because users are required to learn a set of codes or search attributes which must be provided to the system in order to produce desired messages. Thus the rate of communication is limited by the ability of the user to interface with the system, which may ultimately depend on the user's cognitive ability. This requirement is at odds with the basic purpose of assisting those who have difficulty with communication. Moreover, present technology lacks the ability to provide likely subsequent messages based on present communication, and therefore requires complex and often redundant input to invoke each output message. Thus the performance provided by known applications of technology to the goals of ACT is limited by the ability of the user to supply input.
In contrast to present technology, the present invention facilitates communication without requiring the user to learn and to provide complex and redundant input. In general terms, the invention pertains to an ACT system in which user input may specify attributes of the subject of desired communication. Responsive to such input, a system in accordance with the invention may provide to the user a set of messages which are pertinent to the specified subject. From this set of messages the user may specify a message to be produced as output for purposes of communication. Alternatively, the user may alter the specified attributes and receive a different set of messages based on the revised specifications.
The invention may be embodied in a system having a collection of predefined messages for generation as output by the system. The system further includes a predefined organizational structure which associates messages with predefined attributes of predefined subjects of communication. User input specifying predefined attributes may thus be used to produce a set of messages from which the user may select a message for output.
It is preferred that the predefined organizational structure for associating messages with attributes of subjects is provided as a “frame-based” message topology. In this form of message organization, a frame is a set of messages which pertain to a particular subject of communication. Thus a message which may pertain to a variety of subjects, for example, “I feel sore”, will likewise be associated with a frame pertaining to each of those subjects. In the context of the invention, the frame-based message organizational structure is implemented such that predefined subject attributes are associated with predefined frames. Thus upon receiving user specification of subject attributes, the system may provide a frame, i.e. a set of messages, which corresponds to those attributes.
Thus the invention may be embodied in a system for providing augmentative communication in which user specified subject attributes are used to produce a set of messages which are determined to be pertinent to communication as a result of a predefined association with the specified attributes. The invention may similarly be embodied in methods for providing augmentative communication in such a system, and in computer program products for implementing such a system on a computer. Additional features of the invention are disclosed in the figures and detailed description which follow.
REFERENCES:
patent: 4215240 (1980-07-01), Ostrowski
patent: 4785420 (1988-11-01), Little
patent: 5047953 (1991-09-01), Smallwood et al.
patent: 5113481 (1992-05-01), Smallwood et al.
patent: 5204905 (1993-04-01), Mitome
patent: 5210689 (1993-05-01), Baker et al.
patent: 5299125 (1994-03-01), Baker et al.
patent: 5317671 (1994-05-01), Baker et al.
patent: 5345538 (1994-09-01), Narayannan et al.
patent: 5369574 (1994-11-01), Masegi et al.
patent: 5649059 (1997-07-01), Tendler et al.
patent: 5802495 (1998-09-01), Goltra
patent: 5854997 (1998-12-01), Sukeda et al.
patent: 5956667 (1999-09-01), Higginbotham et al.
patent: 5991711 (1999-11-01), Seno et al.
A
Higginbotham D. Jeffery
Wilkins David Parker
Heslin & Rothenberg, P.C.
The Research Foundation of State University of New York
Thomas Joseph
LandOfFree
System and methods for frame-based augmentative... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and methods for frame-based augmentative..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and methods for frame-based augmentative... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2565636