Speech recognition system with changing grammars and grammar help command

• US 6,298,324 B1
• Filed: 11/12/1998
• Issued: 10/02/2001
• Est. Priority Date: 01/05/1998
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Cyr, Craig, Matheson, Lisa, Masters, Steven P., Li, Hang, Zuberec, Sarah E.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): ARMSTRONG, ANGELA A
• Application Number:
  US09/191,048
• Time in Patent Office: 1,055 Days
• Field of Search: 704/260, 704/270, 704/275, 704/276, 704/231, 704/251, 704/254, 704/255
• US Class Current: 704/251
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/225 Feedback of the input speech
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A speech recognition system has an operator interface that helps a user navigate new or changing grammars. The speech recognition system includes an application, a vocabulary that holds a set of utterances applicable to the application, and an active grammar that holds a subset of the utterances in the vocabulary. The active grammar includes a help command as well as other options. The speech recognition system has a speech recognition engine to recognize the utterances in the active grammar, including the help command. When the help command is detected, the operator interface verbally enunciates the utterances in the grammar so that the user can hear the available options in a hands-free, eyes-off operating environment. In this way, any time the user does not know or forgets the available options, he/she simply speaks the help command, such as asking “What can I say.” The system responds verbally with a list of currently available words and/or phrases in the active grammar.

Claims

1. A speech recognition system comprising:

an application;

a vocabulary accessible by the application, the vocabulary including a set of utterances applicable to the application;

an active grammar containing a first subset of utterances from the vocabulary, the first subset comprising one or more keyword utterances, a plurality of non-keyword utterances, and a help command;

a speech recognition engine to recognize the utterances in the active grammar;

an operator interface to;

(a) expand the active grammar to contain both the first subset of utterances and a second subset of utterances from the vocabulary in response to detection by the speech recognition engine of the keyword utterance, wherein the active grammar remains expanded until a non-keyword utterance is detected; and

(b) verbally enunciate the utterances contained in the active grammar in response to recognition by the speech recognition engine of the help command.

2. A speech recognition system as recited in claim 1, wherein the operator interface also lists the utterances on a display in response to recognition of the help command.

3. An operating system that incorporates the speech recognition system as recited in claim 1.

4. A speech recognition system as recited in claim 1, wherein the active grammar remains expanded until both a non-keyword utterance is detected and an operation that corresponds to the non-keyword utterance is complete.

5. A speech recognition system as recited in claim 1, wherein the second subset of utterances contains another keyword and during a third instance following detection of said another keyword, the grammar is expanded a second time to contain the first and second subsets of utterances and a third subset of utterances from the vocabulary.

6. A speech recognition system as recited in claim 5, wherein selected utterances from the first or second subsets of utterances are removed from the doubly expanded grammar.

7. A speech recognition system as recited in claim 1, wherein the one or more keyword utterances do not have an associated function other than to expand the active grammar.

8. A speech recognition system comprising:

an application;

a vocabulary accessible by the application, the vocabulary holding a set of utterances applicable to the application;

an active grammar that holds a first subset of the utterances in the vocabulary, the first subset of utterances including a keyword utterance, a plurality of non-keywords, and a help command;

a speech recognition engine to recognize the utterances in the active grammar, whereupon detection of the keyword utterance causes a change in the active grammar from the first subset of the utterances to a second subset of the utterances, the changed grammar comprising both the first subset of utterances and the second subset of utterances in the vocabulary, the active grammar being reduced to the first subset of utterances upon detection of a non-keyword utterance and upon completion of an operation that corresponds to the non-keyword utterance; and

an operator interface that verbally enunciates to a user the utterances currently contained in the active grammar in response to recondition by the speech recognition engine of the help command.

9. A speech recognition system as recited in claim 8, wherein the operator interface also list the utterances on a display in response to recognition of the help command.

10. An entertainment system incorporating the speech recognition system as recited in claim 8.

11. A computing device incorporating the speech recognition system as recited in claim 8.

12. An operating system incorporating the speech recognition system as recited in claim 8.

13. A speech recognition system as recited in claim 8, wherein the active grammar is reduced to the first subset of utterances upon detection on a non-keyword utterance.

14. A speech recognition system as recited in claim 8, wherein the selected ones of the utterances do not have an associated function other than to expand the active grammar.

15. An operator interface for a speech recognition system, the speech recognition system being configured to recognize a set of utterances, the operator interface comprising:

a vocabulary comprising the set of utterances;

an active grammar that includes a first subset of the utterances in the vocabulary, the first subset of utterances including a help command, one or more keyword utterances, and a plurality of non-keyword utterances;

an audio input that receives audible utterances and passes the utterances to

the speech recognition system for recognition;

an audio speech generator that verbally enunciates the utterances in the active grammar in response to recognition by the speech recognition system of the help command;

wherein the operator interface upon recognition of the keyword utterance temporarily adds a second subset of utterances from the vocabulary to the active grammar so that the active grammar includes both the first and second subset of utterances, and wherein the active grammar is reduced back to include only the first subset of utterances after detection of a non-keyword utterance.

16. An operator interface as recited in claim 15, wherein the audio speech generator comprises:

a text-to-speech converter to create audio data representing the verbal enunciation of the utterances; and

an audio output to play the audio data and verbally enunciate the utterances.

17. A speech recognition system incorporating the operator interface as recited in claim 15.

18. A computing device incorporating the operator interface as recited in claim 15.

19. An operator interface as recited in claim 15, wherein the active grammar remains expanded until both a non-keyword utterance is detected and an operation that corresponds to the non-keyword utterance is complete.

20. An operator interface as recited in claim 15, wherein the one or more keyword utterances do not have an associated function other than to expand the active grammar.

21. A vehicle computer system comprising:

a computer;

an open platform operating system executing on the computer, the operating system being configured to support multiple applications;

a vocabulary comprising a set of utterances;

an active grammar that includes a first subset of the utterances in the vocabulary, the first subset of utterances comprising a help command, a keyword utterance, and a plurality of non-keyword utterances;

a speech recognition system to detect utterances used to control at least one of the applications running on the computer, the keyword utterance, the non-keyword utterances, and the help command utterance, the speech recognition system having an operator interface that;

(a) expands the active grammar to contain both the first subset of utterances and a second subset of utterances from the vocabulary in response to detection by the speech recognition engine of the keyword utterance, wherein the active grammar remains expanded until a non-keyword utterance is detected, and

(b) verbally enunciates to a user the utterances contained in the active grammar in response to detection of the help command utterance.

22. A vehicle computer system as recited in claim 21, wherein the operator interface also lists the utterances on a display in response to recognition of the help command utterance.

23. A vehicle computer system as recited in claim 21, wherein the active grammar remains expanded until both a non-keyword utterance is detected and an operation that corresponds to the non-keyword utterance is complete.

24. A vehicle computer system as recited in claim 21, wherein the expanded active grammar includes a different keyword and whereupon detection of the different keyword, the speech recognition system is configured to expand the active grammar a second time to include the first set of utterances, the second set of utterances, and a third set of utterances.

25. A vehicle computer system as recited in claim 24, wherein the speech recognition system removes selected utterances from the active grammar.

26. A vehicle computer system as recited in claim 21, wherein the keyword utterance does not have an associated function other than to expand the active grammar.

27. In a speech recognition system having an active grammar, the active grammar including a default set of utterances, the default set of utterances including a keyword utterance and a help command, a method for operating the speech recognition system, comprising the following steps:

recognizing the keyword utterance;

in response to recognition of the keyword utterance, ballooning the active grammar to temporarily include both the default set of utterances and a new set of utterances, the active grammar remaining ballooned until a non-keyword utterance is detected, the ballooned active grammar further comprising the non-keyword utterance;

recognizing the help command; and

verbally enunciating the utterances in the active grammar in response to recognition of the help command.

28. A method as recited in claim 27, further comprising reducing the active grammar to the default set of utterances both after recognition of a non-keyword utterance and after completion of an operation that corresponds to the non-keyword utterance.

29. A computer-readable medium as recited in claim 27, wherein the keyword utterance does not have an associated function other than to expand the active grammar.

30. A computer-readable medium comprising instructions for performing steps comprising:

listening for a first predetermined set of utterances that includes a help command, a keyword utterances and a plurality of non-keyword utterances;

in response to detection of the keyword utterance, expanding the predetermined set of utterances to include both the predetermined set of utterances and a second predetermined set of utterances, wherein the utterances remain expanded until detection of a non-keyword utterance;

recognizing the help command; and

in response to recognition of the help command, creating audio data representing verbal enunciation of the utterances in the predetermined set.

31. A computer-readable medium as recited in claim 30, wherein the keyword utterance does not have an associated function other than to expand the active grammar.