IntentRecognizer class
Intent recognizer.
- Extends
Constructors
Intent |
Initializes an instance of the IntentRecognizer. |
Properties
authorization |
Gets the authorization token used to communicate with the service. |
canceled | The event canceled signals that an error occurred during recognition. |
internal |
|
properties | The collection of properties and their values defined for this IntentRecognizer. |
recognized | The event recognized signals that a final recognition result is received. |
recognizing | The event recognizing signals that an intermediate recognition result is received. |
speech |
Gets the spoken language of recognition. |
telemetry |
This method returns the current state of the telemetry setting. |
Inherited Properties
session |
Defines event handler for session started events. |
session |
Defines event handler for session stopped events. |
speech |
Defines event handler for speech stopped events. |
speech |
Defines event handler for speech started events. |
Methods
add |
|
add |
Adds a phrase that should be recognized as intent. |
add |
Adds an intent from Language Understanding service for recognition. |
close(() => void, (error: string) => void) | closes all external resources held by an instance of this class. |
recognize |
Starts intent recognition, and stops after the first utterance is recognized. The task returns the recognition text and intent as result. Note: RecognizeOnceAsync() returns when the first utterance has been recognized, so it is suitable only for single shot recognition like command or query. For long-running recognition, use StartContinuousRecognitionAsync() instead. |
start |
Starts speech recognition, until stopContinuousRecognitionAsync() is called. User must subscribe to events to receive recognition results. |
start |
Starts speech recognition with keyword spotting, until stopKeywordRecognitionAsync() is called. User must subscribe to events to receive recognition results. Note: Key word spotting functionality is only available on the Speech Devices SDK. This functionality is currently not included in the SDK itself. |
stop |
Stops continuous intent recognition. |
stop |
Stops continuous speech recognition. Note: Key word spotting functionality is only available on the Speech Devices SDK. This functionality is currently not included in the SDK itself. |
Inherited Methods
enable |
This method globally enables or disables telemetry. |
Constructor Details
IntentRecognizer(SpeechConfig, AudioConfig)
Initializes an instance of the IntentRecognizer.
new IntentRecognizer(speechConfig: SpeechConfig, audioConfig?: AudioConfig)
Parameters
- speechConfig
- SpeechConfig
The set of configuration properties.
- audioConfig
- AudioConfig
An optional audio input config associated with the recognizer
Property Details
authorizationToken
Gets the authorization token used to communicate with the service.
string authorizationToken
Property Value
string
Authorization token.
canceled
The event canceled signals that an error occurred during recognition.
public canceled: (sender: IntentRecognizer, event: IntentRecognitionCanceledEventArgs) => void
Property Value
(sender: IntentRecognizer, event: IntentRecognitionCanceledEventArgs) => void
internalData
object internalData
Property Value
object
properties
The collection of properties and their values defined for this IntentRecognizer.
PropertyCollection properties
Property Value
The collection of properties and their values defined for this IntentRecognizer.
recognized
The event recognized signals that a final recognition result is received.
public recognized: (sender: IntentRecognizer, event: IntentRecognitionEventArgs) => void
Property Value
(sender: IntentRecognizer, event: IntentRecognitionEventArgs) => void
recognizing
The event recognizing signals that an intermediate recognition result is received.
public recognizing: (sender: IntentRecognizer, event: IntentRecognitionEventArgs) => void
Property Value
(sender: IntentRecognizer, event: IntentRecognitionEventArgs) => void
speechRecognitionLanguage
Gets the spoken language of recognition.
string speechRecognitionLanguage
Property Value
string
the spoken language of recognition.
telemetryEnabled
This method returns the current state of the telemetry setting.
static boolean telemetryEnabled
Property Value
boolean
true if the telemetry is enabled, false otherwise.
Inherited Property Details
sessionStarted
Defines event handler for session started events.
public sessionStarted: (sender: Recognizer, event: SessionEventArgs) => void
Property Value
(sender: Recognizer, event: SessionEventArgs) => void
Inherited From Recognizer.sessionStarted
sessionStopped
Defines event handler for session stopped events.
public sessionStopped: (sender: Recognizer, event: SessionEventArgs) => void
Property Value
(sender: Recognizer, event: SessionEventArgs) => void
Inherited From Recognizer.sessionStopped
speechEndDetected
Defines event handler for speech stopped events.
public speechEndDetected: (sender: Recognizer, event: RecognitionEventArgs) => void
Property Value
(sender: Recognizer, event: RecognitionEventArgs) => void
Inherited From Recognizer.speechEndDetected
speechStartDetected
Defines event handler for speech started events.
public speechStartDetected: (sender: Recognizer, event: RecognitionEventArgs) => void
Property Value
(sender: Recognizer, event: RecognitionEventArgs) => void
Inherited From Recognizer.speechStartDetected
Method Details
addAllIntents(LanguageUnderstandingModel, string)
function addAllIntents(model: LanguageUnderstandingModel, intentId?: string)
Parameters
The language understanding model containing the intents.
- intentId
-
string
A custom id String to be returned in the IntentRecognitionResult's getIntentId() method.
addIntent(string, string)
Adds a phrase that should be recognized as intent.
function addIntent(simplePhrase: string, intentId?: string)
Parameters
- simplePhrase
-
string
- intentId
-
string
A String that represents the identifier of the intent to be recognized.
addIntentWithLanguageModel(string, LanguageUnderstandingModel, string)
Adds an intent from Language Understanding service for recognition.
function addIntentWithLanguageModel(intentId: string, model: LanguageUnderstandingModel, intentName?: string)
Parameters
- intentId
-
string
A String that represents the identifier of the intent to be recognized. Ignored if intentName is empty.
The intent model from Language Understanding service.
- intentName
-
string
The intent name defined in the intent model. If it is empty, all intent names defined in the model will be added.
close(() => void, (error: string) => void)
closes all external resources held by an instance of this class.
function close(cb?: () => void, errorCb?: (error: string) => void)
Parameters
- cb
-
() => void
- errorCb
-
(error: string) => void
recognizeOnceAsync((e: IntentRecognitionResult) => void, (e: string) => void)
Starts intent recognition, and stops after the first utterance is recognized. The task returns the recognition text and intent as result. Note: RecognizeOnceAsync() returns when the first utterance has been recognized, so it is suitable only for single shot recognition like command or query. For long-running recognition, use StartContinuousRecognitionAsync() instead.
function recognizeOnceAsync(cb?: (e: IntentRecognitionResult) => void, err?: (e: string) => void)
Parameters
- cb
-
(e: IntentRecognitionResult) => void
Callback that received the recognition has finished with an IntentRecognitionResult.
- err
-
(e: string) => void
Callback invoked in case of an error.
startContinuousRecognitionAsync(() => void, (e: string) => void)
Starts speech recognition, until stopContinuousRecognitionAsync() is called. User must subscribe to events to receive recognition results.
function startContinuousRecognitionAsync(cb?: () => void, err?: (e: string) => void)
Parameters
- cb
-
() => void
Callback invoked once the recognition has started.
- err
-
(e: string) => void
Callback invoked in case of an error.
startKeywordRecognitionAsync(KeywordRecognitionModel, () => void, (e: string) => void)
Starts speech recognition with keyword spotting, until stopKeywordRecognitionAsync() is called. User must subscribe to events to receive recognition results. Note: Key word spotting functionality is only available on the Speech Devices SDK. This functionality is currently not included in the SDK itself.
function startKeywordRecognitionAsync(model: KeywordRecognitionModel, cb?: () => void, err?: (e: string) => void)
Parameters
- model
- KeywordRecognitionModel
The keyword recognition model that specifies the keyword to be recognized.
- cb
-
() => void
Callback invoked once the recognition has started.
- err
-
(e: string) => void
Callback invoked in case of an error.
stopContinuousRecognitionAsync(() => void, (e: string) => void)
Stops continuous intent recognition.
function stopContinuousRecognitionAsync(cb?: () => void, err?: (e: string) => void)
Parameters
- cb
-
() => void
Callback invoked once the recognition has stopped.
- err
-
(e: string) => void
Callback invoked in case of an error.
stopKeywordRecognitionAsync(() => void, (e: string) => void)
Stops continuous speech recognition. Note: Key word spotting functionality is only available on the Speech Devices SDK. This functionality is currently not included in the SDK itself.
function stopKeywordRecognitionAsync(cb?: () => void, err?: (e: string) => void)
Parameters
- cb
-
() => void
Callback invoked once the recognition has stopped.
- err
-
(e: string) => void
Callback invoked in case of an error.
Inherited Method Details
enableTelemetry(boolean)
This method globally enables or disables telemetry.
static function enableTelemetry(enabled: boolean)
Parameters
- enabled
-
boolean
Global setting for telemetry collection. If set to true, telemetry information like microphone errors, recognition errors are collected and sent to Microsoft. If set to false, no telemetry is sent to Microsoft.
Inherited From Recognizer.enableTelemetry