System and method for isolating and processing common dialog cues
View Patent ↗A method, system and machine-readable medium are provided. Speech input is received at a speech recognition component and recognized output is produced. A common dialog cue from the received speech input or input from a second source is recognized. An action is performed corresponding to the recognized common dialog cue. The performed action includes sending a communication from the speech recognition component to the speech generation component while bypassing a dialog component.
1. A method comprising:
receiving a first speech input from a user interacting with a dialog system at a speech recognition component;
upon recognizing a first cue from the first speech input:
instructing a speech generation component to pause speech generation;
receiving a second speech input from the user; and
while speech generation is paused, developing a potential dialog via a dialog application based on the second speech input; and
upon recognizing a second cue from the second speech input:
instructing the speech generation component to resume the speech generation;
discarding the potential dialog;
recognizing a third speech input using the speech recognition component, to yield data associated with recognized speech from the third speech input; and
outputting the data associated with the recognized speech.
2. The method of claim 1 , wherein the speech recognition component uses a first grammar, the method further comprising:
utilizing a second grammar distinct from the first grammar to recognize the second speech input.
3. The method of claim 2 , wherein the second grammar is configured for improved speech recognition of the second cue.
4. The method of claim 1 , wherein the first cue and the second cue each comprise one of a predefined word and a predefined phrase.
5. The method of claim 1 , wherein pausing speech generation also pauses dialog management.
6. The method of claim 1 , further comprising:
when the first speech input is only partially recognizable, providing a targeted re-prompt.
7. The method of claim 1 , wherein the first cue comprises one of a cough, a sneeze, and a shuffle of paper.
8. A system comprising:
a processor; and
a computer-readable storage medium having instructions stored thereon which, when executed on the processor, cause the processor to perform operations comprising:
receiving a first speech input from a user interacting with a dialog system at a speech recognition component;
upon recognizing a first cue from the first speech input:
instructing a speech generation component to pause speech generation;
receiving a second speech input from the user; and
while speech generation is paused, developing a potential dialog via a dialog application based on the second speech input; and
upon recognizing a second cue from the second speech input:
instructing the speech generation component to resume the speech generation;
discarding the potential dialog;
recognizing a third speech input using the speech recognition component, to yield data associated with recognized speech from the third speech input; and
outputting the data associated with the recognized speech.
9. The system of claim 8 , wherein the speech recognition component uses a first grammar, and wherein the computer-readable storage medium has additional instructions stored which result in the operations further comprising:
utilizing a second grammar distinct from the first grammar to recognize the second speech input.
10. The system of claim 9 , wherein the second grammar is configured for improved speech recognition of the second cue.
11. The system of claim 8 , wherein the first cue and the second cue each comprise one of a predefined word and a predefined phrase.
12. The system of claim 8 , wherein pausing speech generation also pauses dialog management.
13. The system of claim 8 , wherein the computer-readable storage medium has additional instructions stored which result in the operations further comprising:
when the first speech input is only partially recognizable, providing a targeted re-prompt.
14. The system of claim 8 , wherein the first cue comprises one of a cough, a sneeze, and a shuffle of paper.
15. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
receiving a first speech input from a user interacting with a dialog system at a speech recognition component;
upon recognizing a first cue from the first speech input:
instructing a speech generation component to pause speech generation;
receiving a second speech input from the user; and
while speech generation is paused, developing a potential dialog via a dialog application based on the second speech input; and
upon recognizing a second cue from the second speech input:
instructing the speech generation component to resume the speech generation;
discarding the potential dialog;
recognizing a third speech input using the speech recognition component, to yield data associated with recognized speech from the third speech input; and
outputting the data associated with the recognized speech.
16. The computer-readable storage device of claim 15 , wherein the speech recognition component uses a first grammar, and wherein the computer-readable storage medium has additional instructions stored which result in the operations further comprising:
utilizing a second grammar distinct from the first grammar to recognize the second speech input.
17. The computer-readable storage device of claim 16 , wherein the second grammar is configured for improved speech recognition of the second cue.
18. The computer-readable storage device of claim 15 , wherein the first cue and the second cue each comprise one of a predefined word and a predefined phrase.
19. The computer-readable storage device of claim 15 , wherein pausing speech generation also pauses dialog management.
20. The computer-readable storage device of claim 15 , the computer-readable storage device having additional instructions stored which result in the operations further comprising:
when the first speech input is only partially recognizable, providing a targeted re-prompt.