-
Notifications
You must be signed in to change notification settings - Fork 91
Speech
temi's speech flow is comprised of four main components - wakeup, ASR (Automatic Speech Recognition), NLP (Natural language processing) and TTS (text to speech).
temi's SDK provides developers tools to utilize, customize and listen to any of the components.
- To ask temi to speak use the speak(TtsRequest ttsRequest) method and pass a TtsRequest which amongst other attributes will include the text string to be spoken.
- To cancel a TtsRequest use the cancelAllTtsRequests() method.
- To trigger the wakeup programmatically use the method wakeup().
- To get temi's wakeup word use the method wakeWord().
- To listen for tts changes you need to add TtsListener to the addTtsListener(TtsListener ttsListener) method.
- To stop listening for tts changes you need to remove listener using the removeTtsListener(TtsListener ttsListener) method.
- To listen for wakeup triggers you need to add WakeupWordListener to the addWakeupWordListener method.
- To stop listening for wakeup triggers you need to remove the listener using the removeWakeupWordListener method.
- To listen for ASR results you need to add ASRListener to the addAsrListener(ASRListener asrListener) method.
- To stop listening for ASR results you need to remove the listener using the removeAsrListener(ASRListener asrListener) method.
- To let temi actively speak to the user and wait for the user to answer use the askQuestion(String question) method.
- To finish a conversation use the finishConversation() method.
Below are the details regarding the speech methods you can use from SDK.
- Parameters
-
ttsRequest An object of type TtsRequest in this object you will add the text to be spoken - Returns
- void object.
- Exceptions
-
Exception Failed to invoke remote call speak() - Returns
- void object.
- Exceptions
-
Exception Failed to invoke remote call cancelAllTtsRequest() - Returns
- void object.
- Exceptions
-
Exception wakeup() error. - Returns
- Alexa
- Ding Dang
- Hey, temi
- Exceptions
-
Exception getWakeupWord() error - Returns
- void object.
- Exceptions
-
Exception askQuestion() error - Tips
- Returns
- void object.
- Exceptions
-
Exception finishConversation() error
Using this method you are essentially asking temi to add to its' speech request queue another request |
Stops currently processed TTS request and empty the queue |
Use this method to trigger temi's wakeup programmatically |
Use this method to get temi's wake word assistant String object, containing one of the following wakeword options: |
Use this method to let temi actively speak to the user and wait for the user to answer. A custom dialog flow can be realized by cooperating with AsrListener. For details, please refer to Sample code. |
Use this method to finish the conversation (Stop recording for ASR). |
Below are the details regarding how to listen for speech changes.
- Parameters
-
ttsListener A listener object of type TtsListener that contains a void method onTtsStatusChanged(TtsRequest ttsRequest) which contains information regarding a specific speech request - Returns
- void object.
- Parameters
-
ttsListener The tts listener object (of type TtsListener) which you want to stop listening to. - Returns
- void object.
- Parameters
-
wakeupWordListener A listener object of type WakeupWordListener that contains a void method onWakeupWord(String wakeWord) which contains the wakeup word that was used by the user - Returns
- void object.
- Parameters
-
wakeupWordListener The wake word listener object (of type WakeWordListener) which you want to stop listening to. - Returns
- void object.
- Parameters
-
asrListener A listener object of type AsrListener which contains a method onAsrResult(String asrResult) that holds the ASR result in string format - Returns
- void object.
- Tips
- Parameters
- Returns
- void object.
Use this method to add a listener that listens for text to speech status changes. |
Use this method to remove a specific tts listener |
Below are the details regarding how to listen for wakeup triggers.
Use this method to add a listener that will be called when a wakeup word is caught |
Use this method to remove a specific wake word listener |
Use this method to add a listener that will be called when the ASR has returned a result After version 0.10.63, developers can add the following <meta-data> under the <application> tag of AndroidManifest.xml to cover temi’s original voice flow, so developers can access their own NLP service here, and take asrResult as its input: |
Use this method to remove a specific ASR listener
|
Below are the details regarding models used in speech methods
- Attributes
-
id UUID unique number that identifies each tts request speech String The text to be spoken packageName String Skill package name so that temi knows who made the request status Status Status of the request
public enum Status{
PENDING, PROCESSING, STARTED, COMPLETED, ERROR, NOT_ALLOWED
}isShowOnConversationLayer boolean should the conversation line be shown when temi speaks the text
Note: Only relevant for 'Hey temi' assistant skills - Example
- TtsRequest.create("Hi, I'm temi", true);
Request object passed to temi, which contains all the information temi needs to in order to speak and for the skill to track its' request | |||||||||||||||
- Interface
Set your context to implement this listener and add the override method to get tts status changes | ||||||||||||||||||
- Interface
- Parameters
-
wakeupWord String object containing the wakeup word used to trigger direction Which direction did the wakeup come from, or in other words where the user was standing with respect to temi when triggered the wakeup. The direction is returned as an integer and represents the angle of 0 - 360. - 0 - temi was triggered from the front
- 90 - temi was triggered from the left
- 180 - temi was triggered from the back
- 270 - temi was triggered from the right
Set your context to implement this listener and add the override method to get wake word value when triggered by the user |
||||||||||||||||||||||
- Interface
- Parameters
-
asrResult String object containing the words spoken temi after the wakeup trigger
Set your context to implement this listener and add the override method to get the ASR result |
||||||||||||||||||