Speech transcription and analysis system and method

• US 6,714,911 B2
• Filed: 11/15/2001
• Issued: 03/30/2004
• Est. Priority Date: 01/25/2001
• Status: Active Grant

• Current Assignee: Harcourt Assessment Incorporated (Pearson plc)
• Original Assignee: Harcourt Assessment Incorporated (Pearson plc)
• Inventors: Labbe, Laurie, Laurent, Jan C., Waryas, Carol, Parmer, Pam
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): HARPER, V PAUL
• Application Number:
  US09/999,249
• Publication Number:
  US 20020120441A1
• Time in Patent Office: 866 Days
• Field of Search: 706/927, 704/504, 704/503, 704/3, 704/276, 704/271, 704/270, 704/268, 704/267, 704/260, 704/256, 704/254, 704/249, 704/240, 704/231, 704/211, 704/200, 434/362, 434/185
• US Class Current: 704/271
• CPC Class Codes: G09B 19/04 Speaking with audible presentation of the material to be studied G09B5/04
• CPC Class Codes: G09B 5/06 with both visual and audible presentation of the material to be studied
• CPC Class Codes: G10L 2021/0575 Aids for the handicapped in speaking

Abstract

A transcription method uses a computerized process to prompt a student to produce at least one phoneme orally. Next a correct and at least one incorrect production of the phoneme are displayed. The therapist selects from among the displayed productions based upon the student-produced phoneme. The system includes a processor and display to prompt a student to produce at least one phoneme orally, display a correct and at least one incorrect production of the phoneme. The therapist then uses an input device in signal communication with the processor to select from among the displayed correct and incorrect productions based upon the student-produced phoneme, thus obviating the need for the therapist to enter the incorrect production symbol by symbol, unless it is desired to do so, or unless the actual production is not found among the displayed production selections.

Claims

1. A method for use by a therapist to transcribe speech of a student comprising the steps of:

(a) prompting a student to produce a word comprising a plurality of phonemes orally;

(b) displaying a correct production of the word to a therapist;

(c) displaying a phonetic representation of at least one incorrect production of the word to the therapist;

(d) permitting the therapist to make an electronic selection from among the displayed correct and incorrect production representations based upon the student-produced word; and

(e) performing an automatic analysis of a student speech production difficulty based upon an incorrect production representation selection.

2. The method recited in claim 1, wherein the incorrect production representation displaying step comprises displaying a plurality of incorrect production representations.

3. The method recited in claim 2, further comprising the step of, if the student-produced word does not match one of the displayed correct and incorrect production representations, permitting the therapist to enter the student-produced word one phoneme at a time into a processor.

4. The method recited in claim 3, wherein the therapist is permitted to enter each phoneme in International Phonetic Alphabet form.

5. The method recited in claim 1, further comprising the step, following the permitting step, of applying a filter to the incorrect production representation relating to at least one of age and dialect.

6. The method recited in claim 1, further comprising the steps, following the analysis performing step, of collecting a plurality of words from a database based upon the incorrect production representation selection and repeating steps (a)-(d) for each of the collected words.

7. A method for use by a therapist to transcribe speech of a student comprising the steps of:

prompting a student to orally produce a narration comprising a plurality of phonemes by selecting a stimulus for eliciting connected speech;

permitting the therapist to enter the student-produced narration one phoneme at a time into a processor;

performing via software resident in the processor an automatic analysis of a student speech production difficulty based upon at least one incorrect production representation in the entered narration.

8. A system for transcribing speech of a student comprising:

a processor;

display means in signal communication with the processor for;

prompting a student to produce a word comprising a plurality of phonemes orally;

displaying a phonetic representation of a correct production of the at word to a therapist; and

displaying a phonetic representation of at least one incorrect production of the word to the therapist;

input means in signal communication with the processor for permitting the therapist to select from among the displayed phonetic representations of correct and incorrect productions based upon the student produced; and

software means resident on the processor having a code segment adapted to perform an automatic analysis of a student speech production difficulty based upon at least one incorrect production representation in the entered word.

9. The system recited in claim 8, wherein the display means comprises a screen.

10. The system recited in claim 8, wherein the input means comprises a pointing device.

11. A method for use by a therapist to transcribe speech of a student comprising the steps of:

prompting a student to pronounce a word comprising a plurality of phonemes orally;

displaying a phonetic representation of a correct production of the word to a therapist;

displaying a phonetic representation of at least one incorrect production of the word to the therapist;

permitting the therapist to select and perform a method of recording the student-produced word from among the steps of;

making an electronic selection from among the displayed phonetic representations of correct and incorrect productions based upon the student-produced word; and

entering the student-produced word symbolically into a processor; and

automatically performing an analysis of a student speech production difficulty based upon the recorded student-produced word.

12. The method recited in claim 11, wherein the prompting step comprises selecting a stimulus for eliciting connected speech.

13. The method recited in claim 11, wherein the incorrect production phonetic representation displaying step comprises displaying a phonetic representation of a plurality of incorrect productions.

14. The method recited in claim 11, wherein, if the selected recording method comprises entering the student-produced word, permitting each phoneme of the word in International Phonetic Alphabet form.

15. The method recited in claim 12, further comprising the step, following the permitting step, of applying a filter to the phonetic representation of the incorrect production relating to at least one of age and dialect.

16. The method recited in claim 12, further comprising the step, following the analysis performing step, of collecting a plurality of words from a database based upon the phonetic representation of the incorrect production.

17. A system for transcribing speech of a student comprising:

a processor;

display means in signal communication with the processor for;

prompting a student to produce a word orally;

displaying a phonetic representation of a correct production of the word to a therapist; and

displaying a phonetic representation of at least one incorrect production of the word to the therapist;

a first input device in signal communication with the processor adapted to permit the therapist to make an electronic selection from among the displayed correct and incorrect production representations based upon the student-produced word;

a second input device in signal communication with the processor adapted to permit the therapist to enter the student-produced word symbolically into the processor; and

software means resident on the processor having a code segment adapted to perform an automatic analysis of a student speech production difficulty based upon at least one incorrect production representation in the entered word.

18. The system recited in claim 17, wherein the display means comprises a screen.

19. The system recited in claim 17, wherein the first input means comprises a pointing device.

20. The system recited in claim 17, wherein the second input means comprises a keyboard.

21. A method for analyzing a speech problem in a user comprising the steps of:

displaying a symbol to a user, the symbol representative of a word comprising a plurality of phonemes, the displaying step under control of a processor;

prompting a user to pronounce the word orally;

entering the user-pronounced word symbolically into an electronic device;

transmitting the entered symbolic representation of the user-pronounced word to the processor; and

automatically analyzing the symbolic representation of the user-pronounced word to determine whether an error exists in the user pronunciation.

22. The method recited in claim 21, further comprising the step of automatically categorizing an existing error.

23. The method recited in claim 22, wherein the automatic analyzing and categorizing steps are under control of software resident in the processor.

24. The method recited in claim 21, wherein the prompting step comprises selecting a stimulus for eliciting connected speech.

25. The method recited in claim 21, wherein the entering step comprises entering the user-pronounced word in phonetic form.

26. The method recited in claim 21, wherein the analyzing step comprises applying a filter to an incorrect pronunciation relating to at least one of age and dialect.

27. The method recited in claim 21, further comprising the steps of repeating the displaying, prompting, entering, and transmitting steps a plurality of times, and wherein the analyzing step comprises correlating incorrect pronunciations and diagnosing a problem speech sound from the correlation.

28. Computer software for analyzing a speech problem in a user comprising:

a code segment for displaying a symbol to a user on a display device, the symbol representative of a word comprising a plurality of phonemes;

a code segment for prompting a user to pronounce the word orally;

a code segment for receiving a symbolic representation of the user-pronounced word; and

a code segment for analyzing the symbolic representation of the user-pronounced word to determine whether an error exists in the user pronunciation.

29. The computer software recited in claim 28, further comprising a code segment for automatically categorizing an existing error.

30. The computer software recited in claim 28, wherein symbolic representation of the user-pronounced word comprises a phonetic form.

31. The computer software recited in claim 28, wherein the analyzing code segment includes a filter for application to an incorrect pronunciation relating to at least one of age and dialect.

32. The computer software recited in claim 28, further comprising a code segment for directing a repetition of the code segments for displaying, prompting, entering, and transmitting a plurality of times, and wherein the analyzing code segment comprises a code segment for correlating incorrect pronunciations and diagnosing a problem speech sound from the correlation.

33. A system for analyzing a speech problem in a user comprising:

a processor;

a display device;

a therapist input device adapted to receive a symbolic representation of a word comprising a plurality of phonemes, the therapist input device comprising means for transmitting data to the processor;

software means resident on the processor for;

displaying a symbol to a user on the display device, the symbol representative of a word comprising a plurality of phonemes;

prompting a user to pronounce the word orally;

receiving a symbolic representation of the user-pronounced word from the therapist input device; and

automatically analyzing the symbolic representation of the user-pronounced word to determine whether an error exists in the user pronunciation.

34. The system recited in claim 33, wherein the software means is further adapted to automatically categorize an existing error.

35. The system recited in claim 33, wherein symbolic representation of the user-pronounced word comprises a phonetic form.

36. The system recited in claim 33, wherein the analyzing portion of the software means includes a filter for application to an incorrect pronunciation relating to at least one of age and dialect.

37. The system recited in claim 33, wherein the software means is further adapted to direct a repetition of the displaying, prompting, entering, and transmitting a plurality of times, to correlate incorrect pronunciations, and to diagnose a problem speech sound from the correlation.