​
Login / Sign Up

  1. Go to SubVerse Dashboard
  2. Sign up or log in with your credentials.

​
Create a Voice Agent

  1. Click on Create New Agent or select a template from the list.
  2. Provide a Usecase Name (only alphanumeric and underscores are allowed).

To save time, you can also duplicate any existing use case by clicking on the copy icon.

​
Customize your Agent

  • Initial Prompt: Set an introductory message for the agent.
  • System Prompt: Define agent behavior, identity, and response style.
  • Speech Detection Settings: Choose language, interruptions, and accuracy.
  • Voice Settings: Select TTS language, voice, quality, and speed.
  • Call Settings: Configure transfer numbers (for realtime escalations to human agents), communication methods, and DTMF settings.
  • LLM Tools: Select tasks like end_call, transfer_call, or query_knowledge_base.
  • Knowledge Base: Attach relevant company documents for agent reference.

​
Test Call

  1. Once you save the agent configuration, hover over to the voice agent you just created.
  2. Click on Make Call icon to initiate a test conversation.
  3. Adjust settings as needed and refine the agent responses.

Agent Configuration

Best practices for agent prompting and other settings

Telephony

Setup telephony to make or receive calls

Workflow

Take more actions by setting up workflow automation

Monitor Performance

See transcripts, recordings, analysis of agent-customer interactions