In this tutorial, we’ll set up a simple callback event that includes response data and allows you to mark whether the initial response data would be used as final response data.

Prerequisites

Before proceeding with this tutorial, complete the setup steps outlined in Voice SDK tutorials to create a project with the necessary dependencies. This tutorial builds upon that project.

Live Understanding Short Response Tutorial

Step 1: Train your Wit.ai application for the intent and entity.

In your new Wit.ai application, open the Understanding page.

Under Utterance enter Set to blue.

Next to Intent, enter set_color, and then click + Create Intent.

In the Utterance box, highlight the word blue.

In the entity box, enter color1, and then click + Create Entity.

Click Train and Validate. "Training and Validating Your Utterance"

Repeat this process for other utterances that are shorter and include color names (including this one), such as make blue, turn red, green, purple, orange, and other colors, including white, black, or gray. Repeat this process at least 10 to 15 times.

Note Ensure that you select the set_color intent and mark the color itself as a color entity.

Go to Management > Settings and copy the Server Access Token for later use.

Step 2: Creating your scene

In the Hierarchy window of a blank scene in your new Unity project, select + > 3D Object and select Cube. "Adding a 3D Cube"

Go to + > UI > Button - TextMeshPro. Import TextMeshPro Essentials for the labels. Name the button Activate Button.

Expand the button and change the name Text (TMP) to Activation Label.

Go to + > UI > Text - TextmeshPro, add a text object and call it Transcription Label.

Go to Assets > Create > Voice SDK and select Add App Voice Experience to scene. "Adding App Voice Experience to a Scene"

Select the button created and click the + beneath On Click (). Drag the App Voice Experience GameObject onto the blank object to add it to the On Click () method.

Open the AppVoiceExperience.Activate() method that you just dropped into the button, and then expand Events.

Under Activation/Deactivation Events, click the + beneath On Start Listening (), and then drag the button label object onto the blank object to add it to the event.

From the list next to Runtime Only, select TextMeshProGUI > string text, and then enter Deactivate in the box. "Setting event to deactivate"

Repeat the last two steps for the On Stopped Listening () event and enter Activate.

Under Transcription Events, click the + beneath On Partial Transcription (string), and then drag the Transcription Label object onto the blank to add it to the event. From the list next to Runtime Only, select to TextMeshProGUI > text, then do the same for On Full Transcription (string) "Setting Transcription Events"

Under Activation/Deactivation Events, click the + beneath On Start Listening (), and then drag the Transcription Label object onto the blank to add it to the event, so the transcription label will reset to an empty string.

Step 3: Integrating Voice SDK with your app

Copy the Server Access Token from your Wit.ai application.

In the Unity Editor, go to Meta > Voice SDK > Settings and paste in the Server Access Token. "Setting Server Access Token"

Click Link and save your configuration in the Assets directory as ShortResponseWitConfig.

On the AppVoiceExperience GameObject, drag your newly generated ShortResponseWitConfig onto the Wit Configuration field.

Test your app by clicking the play button at the top of the Unity editor. Hit the Activate button in your app to begin speaking and make sure that a transcription of what you say is reflected in the transcription label.

Did you find this page helpful?