AI configuration allows you to customize the AI moderator's language, voice, avatar, speaking, listening and silence settings, to perfect it as you like.
Accessing the AI Configuration Feature
To access the AI configuration feature, navigate to the Study settings from left nav bar. Here, you will find the AI configuration tab.
Modifying Avatar Library and Voice Settings
In the AI configuration tab, you can modify the avatar library and adjust various voice settings such as age, gender, and accents.
Understanding the AI Configuration Tabs
The AI configuration feature comprises three tabs: Speaking, Listening, and Silence. Each tab controls different aspects of the AI moderator's behavior.
Speaking Tab
The Speaking tab includes settings for:
-
Speech normalization - Setting for speaking out the number out loud or not. For example - 22 will be read as "twenty two" with this setting enabled.
-
Responsiveness - For controlling how responsive the AI moderator. Smaller numbers mean longer wait times for AI to speak after the participants response and vice versa.
-
Speaking speed - For controlling the speed with which the AI moderator speaks.
-
Interruption sensitivity - Controls how easily the AI moderator will be interrupted by the participants responses.
-
Voice emotional intensity - Controls the tonality in the voice of the AI moderator.
-
The AI's first message - You can set an opening message for starting the interview with.
![Image]({"zoomFactor":1,"shapes":[],"s3ObjectId":"b62793f5-64fa-4023-a588-901dd86b5ae6.png","timestamp":81.94,"stepId":"b62793f5-64fa-4023-a588-901dd86b5ae6"})
Listening Tab
The Listening tab allows you to:
-
Select input languages
-
Specify expected keywords
These keywords help the AI moderator recognize new terminologies or proper nouns used in your study.
![Image]({"stepId":"d5b01ce4-7313-4341-a7bc-258d6c694d89","zoomCenter":{"x":0.6780202235231506,"y":0.6157205240174672},"naturalDimensions":{"x":1879,"y":916},"zoomFactor":2,"shapes":[],"timestamp":104.633,"s3ObjectId":"e26a0e31-4048-4c54-9dea-511ccdc53f7d.png"})
Silence Tab
The Silence tab includes settings for:
-
Think out loud nudges - You can set the duration after which the AI moderator will prompt participants to speak if they remain silent for a specified period.
-
End call settings - Set a duration for ending the call after a specified time limit.
-
Active listening - Use responses like "yeah", "hmm", etc to acknowledge participant responses during the meetings.
-
Response frequency - Set how frequently do you want the AI moderator to use active listening responses.
-
Acknowledgement responses - Set the exact words to be used for active listening responses.
These settings help maintain a smooth conversation with the participants.
Finalizing Your Settings
Once you have configured all the settings, you can check if these changes are reflected in the AI voice by clicking on "Start Conversation" and talking with AI. Iterate until you're satisfied with your AI configuration.
If you need further help, please email us at support@userology.co