org.myrobotlab.service.interfaces

Interface SpeechRecognizer

All Superinterfaces:

NameProvider

All Known Implementing Classes:

AbstractSpeechRecognizer, AndroidSpeechRecognition, Sphinx, WebkitSpeechRecognition

public interface SpeechRecognizer
extends NameProvider

Method Summary

Modifier and Type	Method and Description
void	addMouth(SpeechSynthesis mouth) This method should listen for Mouth events FIXME - should be deprecated - use Attach Pattern !
void	clearLock() This will unlock lockOutAllGrammarExcept(lockPhrase)
boolean	isListening() track the state of listening process
void	listeningEvent(java.lang.Boolean event) Event is sent when the listening Service is actually listening or not.
void	lockOutAllGrammarExcept(java.lang.String lockPhrase) The ear service will not listen anymore until the magical keyword "lockPhrase" said or clearLock() method called
void	onEndSpeaking(java.lang.String utterance)
void	onStartSpeaking(java.lang.String utterance)
void	pauseListening() method to suppress recognition listening events This is important when a Speech Recognizer is listening --> then Speaking, typically you don't want the STT to listen to its own speech, it causes a feedback loop and with STT not really very accurate, it leads to weirdness -- additionally it does not recreate the speech processor - so its not as heavy handed
java.lang.String	recognized(java.lang.String word)
void	resumeListening()
void	setAutoListen(boolean autoListen) If setAutoListen is True, webkitspeech red microphone will auto rearm.
void	startListening()
void	stopListening()

Methods inherited from interface org.myrobotlab.framework.interfaces.NameProvider

getName

Method Detail

listeningEvent

void listeningEvent(java.lang.Boolean event)

Event is sent when the listening Service is actually listening or not.

pauseListening

void pauseListening()

method to suppress recognition listening events This is important when a Speech Recognizer is listening --> then Speaking, typically you don't want the STT to listen to its own speech, it causes a feedback loop and with STT not really very accurate, it leads to weirdness -- additionally it does not recreate the speech processor - so its not as heavy handed

recognized

java.lang.String recognized(java.lang.String word)

resumeListening

void resumeListening()

startListening

void startListening()

stopListening

void stopListening()

addMouth

void addMouth(SpeechSynthesis mouth)

This method should listen for Mouth events FIXME - should be deprecated - use Attach Pattern !

onStartSpeaking

void onStartSpeaking(java.lang.String utterance)

onEndSpeaking

void onEndSpeaking(java.lang.String utterance)

lockOutAllGrammarExcept

void lockOutAllGrammarExcept(java.lang.String lockPhrase)

The ear service will not listen anymore until the magical keyword "lockPhrase" said or clearLock() method called

clearLock

void clearLock()

This will unlock lockOutAllGrammarExcept(lockPhrase)

setAutoListen

void setAutoListen(boolean autoListen)

If setAutoListen is True, webkitspeech red microphone will auto rearm. microphone will shutdown too if mouth is activated. Careful if this is set to True : You cannot control anymore red microphone from webgui You need to control it from SwinGui, or usually from code

isListening

boolean isListening()

track the state of listening process