Creating a Project

This section explains how to create a Digital Person experience. First, you could create a basic Digital Person using our sample conversation and default UI. You then must determine which Digital Person behavior, skills, features, language, conversation provider, etc., are most relevant to your use case and fine-tune your project accordingly. The preview in the right pane will show you the changes that have been applied allowing you to test each configuration before deploying the project. Finally, when you are satisfied with your Digital Person, deploy it, and share the URL with the end-users.

Please note: Every time you make a change to the configuration, you must click Update Preview on the Preview window to incorporate the changes into the preview DP’s behavior.

Please click on the expand link below each step for additional information about the configuration.

Video Tutorial

As our product has improved rapidly, the video above maybe outdated, but the sections below are up to date.

Prerequisites

Before you begin creating your project, ensure that you have:

Access to a corpus. We support integrations with Watson, Dialogflow, and Azure Bot Service. We also support other NLP software platforms with a little extra effort.

Complete the following steps to create a new Digital Person project.

1. Choose a name for your project

Click Create a New Project and then Name the project.
It must be a unique and meaningful name as this name will also be in your deploy links. The system does not allow you to use a name that has been already used for an existing project.

2. Select your Digital Person avatar

Available to all our customers are the latest HumanOS2.6 Digital People. If you have built your own Custom Digital People, they will also be available to you.

a8013c8d-d597-4117-a378-82d0dfffea65 (1).png

If you would like to use any of the Digital Persons as a starting point and change the look, click on the 3 dots within the profile picture and select Collect Digital DNA. This gives you the unique blend used to create this Digital Person in an encoded format (a.k.a Digital DNA). Copy the DNA and save it in your preferred text editor.

To continue working on the Digital Person, use the Loading Digital DNA feature of the Digital DNA Blender.

Note: The Collect Digital DNA option is available only for the Digital People created from Digital DNA Blender.

3. Choose the behavior of your Digital Person

Some of the behavior options in Digital DNA studio are dependent on Human OS versions, and languages. Refer to the Feature Matrixarchived and Features Supported on Languages for details.

Choose a behavior style

Behaviour styles are a predefined set of behaviors consisting of defined gestural and emotional behavior, that can be easily matched to the written persona you’ve created and your use case.

Behavior adjustments

Adjust behavior with sentence or word based behavior tags

Other behavioral settings

Boost expressiveness with additional iconic gestures - You can choose whether or not to automatically insert additional gestures. Turning this on will create a more expressive digital person. If you want a digital person to show less hand gestures you can turn this toggle off.

Enable Digital Person Greeting at the start - This option makes your Digital Person greet users as it starts up to provide them with a verbal cue that the interaction has commenced. This setting is enabled by default.

End Session Timeout - This option makes your Digital Person automatically end the conversation session with the user, after a specified period of inactivity from the user side, e.g. after 5 minutes. This setting is enabled by default.

React to negative facial expressions - You can choose whether or not to allow your digital person to respond to users' negative facial expressions.

React to negative speech - You can choose whether or not to allow your digital person to respond to users' negative speech.

Please refer to this page for more details on Behavior Adjustments and other behavioral settings.

Choose a behavior style for your digital person

Behavior adjustments and other behavioral settings

4. Connect conversation and add skills

Skills are modular components that can be added to the Digital Person to enhance their conversation capabilities. See Adding Skills in your Digital Person for detailed information on adding and managing Skills.

Notes:

Dialogflow CX NLP is available only for HumanOS 2.2 Digital People.
Skills can only be used with HumanOS 2.2 Digital People and the Supported NLP platforms.
If you’d like to use an NLP that's not listed here, follow the instructions described in the Orchestration Server Configuration Guide and skip the Manage Conversation and Skills step.

Click Browse Options.
Choose one of the base conversations from the list:
- Sample Conversation enables you to quickly experience and test a Digital Person and Skills through a default pre-configured NLP—you do not need to set NLP accounts, create content, or configure credentials.
- For Watson, click edit and browse for the downloaded credentials file.
- For Dialogflow ES, click edit, and browse for the downloaded credentials file.
- For Dialogflow CX, click edit, and browse for the downloaded credentials file.
- For Azure Bot Service, copy the Microsoft Bot Direct Line API 3.0 secret key from the Azure Portal and paste it into the dialog box.
- For Amazon Lex, enter the AWS role ID and Lex Bot ID into the Amazon Lex configuration screen.
- For Kore AI, enter the Kore AI botID, clientID, and clientSecret into the Kore AI NLP configuration screen.
If the credentials file is valid, a preview appears and you can click Enable As Base. You can then add skills or complete the conversation setup.
Click Add Skills and enable the Skills you want your Digital Person to have (optional). This button is activated only when you choose the supported NLP.
Click Finish Setup.

Video- Add Base Corpus and Skills

5. Choose a language for the conversation

Choose your Digital Person’s language.
This is the language in which your Digital Person will speak. By default, English (US) is selected.
Choose a voice and accent.
We support TTS from Google, Microsoft, and Amazon.
To test the voice
1. click Update preview on the Preview window
2. open the text entry window
3. Then type in a test sentence and click Repeat.

Choose Audience language.
This is the language your target audience speaks, which the Digital Person understands.
Add Speech Hints.
If using Google Speech Services, this feature enables you to build a list of uncommon words or phrases that you want your Digital Person to understand. Type in commonly misinterpreted words or phrases that you want your Digital Person to be trained to recognize and understand.

6. Customize your UI

Your UI is the background and set of controls that your Digital Person lives within.

Add a logo and customize colors for the UI (optional).

7. Customise camera behavior

Note: Camera behavior option is only available only for Human OS 2.2 Digital Persons and later.

Content and cinematic cuts can enhance the experience of interacting with a Digital Person by introducing a range of server-controlled camera cuts that respond to relevant objects appearing on screen. Choose one of the below camera behavior to configure the Digital Person’s interaction when content cards are displayed on the UI. For more information on content cards, see Displaying Content in the Digital DNA Studio Default UI.

Click the See preview link to view a short preview of the interaction.

8. Configure an orchestration server (optional)

Use an Orchestration Server if you're using unsupported NLP services, building a custom backend, implementing rich authentication, or requiring flexible data analytics. Orchestration servers are useful when you are using unsupported NLP services, designing a custom backend, implementing rich authentication, or need flexible data analytics.

To quickly test our project you can skip this step by choosing Sample Conversation in step 3.

9. Preview your settings

The preview pane on the right hand side is a handy tool to test your configurations allowing you to fine tune your Digital Person. Use the text box to write your own sample script and test the voice and behaviour settings. Here’s a list of features you can test in the preview:

Changing avatar
Changing TTS/voice
Changing camera
Changing behaviour/personality
Gesture Markup

Every time you change one of those settings, click Update Preview on the Preview window to incorporate the changes into the preview DP’s behavior.

To show the text box and insert your own text that the DP can repeat:

Open the text entry window
Then type in a test sentence and click Repeat.

10. Deploy your project

Click Save to save your work.
You can return it at any time and pick it up from where you left off.
Click Deploy to apply and view the new changes to your project.
You are presented with a URL to visit where your demo has been deployed.

Note: Projects take between 15-30 seconds to launch the experience for the first time after deploying. After selecting Deploy Now the deploy process will begin and you will be redirected to the Project Management page.

Deploy Project

Soul Machines Knowledge Base