Speech recognition system which displays a subject for recognizing an inputted voice

• US 6,374,218 B2
• Filed: 02/20/1998
• Issued: 04/16/2002
• Est. Priority Date: 08/08/1997
• Status: Expired due to Term

• Current Assignee: Eloqui Voice Systems, LLC (2S Ventures, LLC)
• Original Assignee: Fujitsu Limited
• Inventors: Satoh, Yasuo, Koide, Tomio, Kimura, Shinta
• Primary Examiner(s): Chawan, Vijay B
• Application Number:
  US09/026,612
• Publication Number:
  US 20010012995A1
• Time in Patent Office: 1,516 Days
• Field of Search: 704/251, 704/231, 704/270, 704/260, 704/225, 704/200, 707/500, 707/531
• US Class Current: 704/251
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/225 Feedback of the input speech

Abstract

A speech recognition system wherein a word being the most similar to an inputted voice is delivered as a recognized result; in order to assist a user in entering a voice input, words being subjects for the recognition of the inputted voice are automatically displayed, or recently displayed ones of words which have been displayed up to the present are extracted as words to become subjects for recognition and are automatically displayed as the subjects for the recognition of the input voice.

Claims

1. A speech recognition system for recognizing an inputted voice command of a user, comprising:

a recognitive word storage unit to store recognitive subject words which are subjects for the recognition of the inputted voice command;

a display unit to automatically display and scroll the words stored in said recognitive word storage unit subject to recognition, before the user first inputs a voice command, such that the user finds a word to be entered with the input voice command without entering a computer mouse input;

an input unit to input a voice command of a user; and

a speech recognition unit to recognize a word which is most similar to said inputted voice command from among recently displayed words from the recognitive subject words stored in said recognitive-word storage unit, on the basis of a feature of said inputted voice command, so as to deliver the recognized word as a recognized result.

2. A speech recognition system as defined in claim 1, wherein said display unit displays said recognitive subject words stored in said recognitive word storage unit, while automatically scrolling said words on a display screen thereof.

3. A speech recognition system as defined in claim 2, wherein said display unit determines timings of the scroll on the basis of a clock signal.

4. A speech recognition system as defined in claim 1, wherein said display unit displays said recognitive subject words stored in said recognitive word storage unit, while automatically repaginating said words on a display screen thereof.

5. A speech recognition system as defined in claim 4, wherein said display means determines timings of the repagination on the basis of a clock signal.

6. A speech recognition system as defined in claim 1, wherein said display unit reads out said recognitive subject words stored in said recognitive word storage unit, successively in units of at least one word, so as to automatically display said at least one word on a display screen thereof.

7. A speech recognition system as defined in claim 6, wherein said display unit determines timings of the successive readout on the basis of a clock signal.

8. A speech recognition system as defined in claim 1, wherein said display unit reads out said recognitive subject words stored in said recognitive word storage unit, in units of at least one word on the basis of values of random numbers which correspond to numbers assigned to said words, so as to automatically display said at least one word on a display screen thereof.

9. A speech recognition system for recognizing an inputted voice command, comprising:

a recognitive word storage unit to store words which are subjects for the recognition of the currently inputted voice command;

a word storage unit to store recognitive subject words which are subjects for the recognition by the recognition system;

a display unit to automatically display and scroll the words stored in said word storage unit before a user first speaks the voice command, in order to assist the user of said speech recognition system in entering the voice command;

an input unit to input a voice command of a user;

a speech recognition unit to recognize a word which is the most similar to said inputted voice command from among recently displayed words stored in said recognitive word storage unit, on the basis of a feature of said inputted voice command, so as to deliver the recognized word as a recognized result; and

a display-word extraction unit to extract words recently displayed by said display unit, and to store the extracted words in said recognitive word storage unit.

10. A speech recognition system as defined in claim 9, wherein said display unit displays said recognitive subject words stored in said recognitive word storage unit, while automatically scrolling said words on a display screen thereof.

11. A speech recognition system as defined in claim 10, wherein said display unit determines timings of the scroll on the basis of a clock signal.

12. A speech recognition system as defined in claim 9, wherein said display unit displays said recognitive subject words stored in said recognitive word storage unit, while automatically repaginating said words on a display screen thereof.

13. A speech recognition system as defined in claim 12, wherein said display unit determines timings of the repagination on the basis of a clock signal.

14. A speech recognition system as defined in claim 9, wherein said display unit reads out said recognitive subject words stored in said recognitive word storage unit, successively in units of at least one word, so as to automatically display said at least one word on a display screen thereof.

15. A speech recognition system as defined in claim 14, wherein said display unit determines timings of the successive readout on the basis of a clock signal.

16. A speech recognition system as defined in claim 9, wherein said display unit reads out said recognitive subject words stored in said recognitive word storage unit, in units of at least one word on the basis of values of random numbers which correspond to numbers assigned to said words, so as to automatically display said at least one word on a display screen thereof.

17. A computer-readable storage medium storing therein a program for causing a computer to perform:

automatically displaying and scrolling words stored in a recognitive-word storage unit subject to recognition, before a user first inputs the voice command, such that a user finds a word to be entered with the input voice command without entering a computer mouse input; and

obtaining a word which is most similar to the inputted voice command from among recently displayed words that are stored in the recognitive-word storage unit, on the basis of a feature of the inputted voice so as to deliver a most similar word as a recognized result.

18. A computer-readable storage medium storing therein a program for causing a computer to execute, comprising:

automatically displaying and scrolling words stored in a word storage unit storing the words which may possibly become subjects for recognition of an input voice command, before a user first speaks the inputted voice command, such that a user finds a word to be entered with a voice input without entering a computer mouse input;

extracting words recently displayed among the words which have been automatically displayed up to the present, as the words to become said recognitive subject words, so as to store the extracted words in the recognitive-word storage unit; and

obtaining a word which is most similar to an inputted voice command from among the extracted words stored in a recognitive word storage unit storing the recognitive subject words for the recognition of the currently inputted voice command, on the basis of a feature of said inputted voice command, so as to deliver a most similar word as a recognized result.

19. A method for displaying a recognitive-subject, in a speech recognition system for recognizing an inputted voice command, comprising:

storing words which are subjects for recognition by the speech recognition system;

automatically displaying and scrolling said stored words before a user first speaks the voice command, such that the user finds a word to be entered with the voice command without entering a computer mouse input;

inputting the voice command of a user; and

obtaining a word which is most similar to said inputted voice command from among recently displayed word from the stored recognitive subject words, on the basis of a feature of said inputted voice command, so as to deliver a most similar word as a recognized result.

20. A method for displaying a recognitive-subject, in a speech recognition system for recognizing an inputted voice, comprising:

automatically displaying and scrolling, before a user first speaks the voice command, words stored in a word storage unit storing the words which may possibly become subjects for recognition of input voice commands, such that a user finds a word to be entered with an input voice command without entering a computer mouse input;

inputting the voice command of the user;

obtaining a word which is most similar to an inputted voice command from among recently displayed words from the recognitive subject words being subjects for recognition of the currently inputted voice command, on the basis of a feature of said inputted voice command, so as to deliver a most similar word as a recognized result; and

extracting words recently displayed among words which have been displayed up to the present by said displaying step, as the words to become the recognitive subject words, so as to store the extracted words in a recognitive-word storage unit.

21. A speech recognition system, comprising:

a storage unit to store recognizable words;

a display unit to automatically display and scroll, before a user first speaks a voice command word, at least one of the recognizable words in a predetermined sequence on a display prior to an audible input of the voice command word to a speech recognition unit; and

a input unit to input the voice command of a user,

wherein said speech recognition unit recognizes the inputted voice command word by selecting at least one of the recognized words from among the recently displayed words from said storage unit which is most similar to the inputted voice command word such that the user finds a word to be entered with the voice command word without entering a computer mouse input.

22. A speech recognition system for recognizing an inputted voice command, comprising:

a recognitive word storage unit to store recognitive subject words which are subjects for the recognition of the inputted voice command;

a display unit to automatically scroll the words stored in said recognitive word storage unit across a display, before a user speaks the inputted voice command, on the basis of a clock signal;

an input unit to input the voice command of a user; and

a speech recognition unit to recognize a word which is most similar to said inputted voice command from among recently displayed words stored in said recognitive-word storage unit, on the basis of a feature of said inputted voice command, so as to deliver the recognized word as a recognized result.

23. A speech recognition system for recognizing an inputted voice command, comprising:

a recognitive word storage unit to store recognitive subject words which are subjects for the recognition of the inputted voice command;

a display unit reads out said recognitive subject words stored in said recognitive word storage unit, in units of at least one word on the basis of values of random numbers which correspond to numbers assigned to said words, to automatically display and scroll said at least one word on a display, before a user first inputs the voice command;

an input unit to input the voice command of the user; and

a speech recognition unit to recognize a word which is most similar to said inputted voice command from among recently displayed subject words stored in said recognitive-word storage unit, on the basis of a feature of said inputted voice, so as to deliver the recognized word as a recognized result.

24. A speech recognition system that recognizes an audible input as a recognized word, comprising:

a display unit to sequentially display and scroll a plurality of stored recognizable words, before a user first speaks the audible input, such that each word is displayed without displaying other words of the plurality;

an input unit to input the audible input of a user; and

a selection unit to select a recently displayed word as a recognized word based on the audible input.