System for synthesizing spoken messages

• US 6,068,485 A
• Filed: 05/01/1998
• Issued: 05/30/2000
• Est. Priority Date: 05/01/1998
• Status: Expired due to Term

• Current Assignee: Unisys Corporation
• Original Assignee: Unisys Corporation
• Inventors: Romania, John F., Linebarger, Marcia C.
• Primary Examiner(s): Hafer, Robert A.
• Assistant Examiner(s): Rovnak, John Edmund
• Application Number:
  US09/071,472
• Time in Patent Office: 760 Days
• Field of Search: 434/112, 434/116, 434/156, 434/167, 434/169, 434/176, 434/185, 434/307 R, 434/308, 434/309, 434/319, 704/270, 704/271
• US Class Current: 434/116
• CPC Class Codes: G09B 19/04 Speaking with audible presentation of the material to be studied G09B5/04

Abstract

A computer-operated system for assisting aphasics in communication includes user-controlled means for storing data representing the user'"'"'s vocalizations during a time interval, means for associating the data stored in each of a plurality of such intervals with an icon, means for ordering a plurality of such icons in a group representing a speech message, and means for generating an audio output from the stored data represented by the icons in the group so as to provide a speech message.

Claims

1. A computer-implemented method for enabling a patient to generate a speech message, the method comprising the steps of:

recording a plurality of digital data files, wherein the step of recording is performed by the patient, and each one of the digital data files representing a respective speech segment spoken by the patient during a respective time interval;

associating a different icon with each one of the digital data files;

arranging the icons in an order representing the speech message composed of the respective speech segments; and

generating an audible speech message from the digital data files in the order represented by the icons.

2. The method of claim 1, further comprising the step of e-mailing the digital data files representing the speech message.

3. The method of claim 1, further comprising the step of performing speech recognition on the digital data files representing the speech message to provide text files representing the speech message.

4. The method of claim 3, further comprising the step of performing natural language processing on the text files representing the speech message to determine the grammaticality of the speech message.

5. The method of claim 1, wherein the step of associating includes associating a different arbitrary icon with each one of the digital data files.

6. The method of claim 1, further comprising the step of evaluating a selected one of the digital data files, and further comprising the step of re-recording the selected digital data file based on a result of the evaluating step.

7. A computer-implemented method of enabling a speech-impaired patient to synthesize a speech message, the method comprising the steps of:

recording at least first and second speech elements spoken by the patient, wherein the step of recording is performed by the patient;

digitizing the first and second speech elements to provide respective first and second digital representations corresponding to the first and second speech elements;

storing the first and second digital representations;

associating first and second icons respectively with the first and second digital representations;

arranging the first and second icons to form the speech message; and

generating an audio output by playing the first and second digital representations.

8. The method of claim 7, wherein the step of associating includes associating a different arbitrary icon with each one of the digital representations.

9. The method of claim 7, wherein the step of arranging includes placing the first and second icons into a sentence assembly area in an order, and wherein the step of generating includes playing the digital representations corresponding to the first and second icons.

10. The method of claim 7, wherein the step of generating an audio output includes playing the digital representations corresponding to the first and second icons placed in the sentence assembly area.

11. The method of claim 7, further comprising the steps of grouping the first and the second icons and placing the first and the second icons as a group into a paragraph assembly area.

12. The method of claim 11, wherein the step of generating an audio output includes playing the digital representation corresponding to the first and second icons placed in the paragraph assembly area.

13. The method of claim 7, further comprising the step of evaluating a selected one of the speech elements, and the step of re-recording the at least first and second speech elements spoken by the patient based on a result of the evaluating step.

14. The method of claim 7, further comprising the step of performing speech recognition on the audio output to produce a text output.

15. The method of claim 14, further comprising the step of performing natural language processing on the text output.

16. The method of claim 15, further comprising the step of performing grammaticality analysis on the text output.

17. The method of claim 14, further comprising the step of e-mailing the audio output.

18. A computer implemented interface enabling a patient to synthesize a speech message, the interface comprising:

means for recording a plurality of speech segments spoken by the patient;

means for associating each one of the speech segments with a respective sound icon that is responsive to patient input to output the speech segment associated with the icon;

an input workspace area adapted to store the sound icons;

a sentence assembly area including a plurality of word holding icons, each one of the word holding icons adapted to contain one of the sound icons, the sentence assembly area being responsive to patient input to group each one of the sound icons contained in the sentence assembly area into a sentence representing sound icon, thereby synthesizing at least a portion of the speech message;

means, responsive to patient input, for playing a speech element associated with one each of the sound icons contained in the sentence assembly area;

a paragraph assembly area including a plurality of sentence holding icons, each one of the sentence holding icons adapted to contain one of the sentence representing sound icons; and

means, responsive to patient input, for playing the speech elements associated with each of the sentence representing sound icons contained in the sentence assembly area, thereby outputting at least a portion of the speech message to the patient.

19. The interface of claim 18, further comprising means for e-mailing at least a portion of the speech message.