Skip to main content
DialogBot blends AI voice recognition with LLM-based replies to conduct live phone conversations. Here is the simplified flow from the moment a call connects to the moment it ends.

​
The call flow

1

A call is initiated

  • Inbound: A customer dials your assigned phone number.
  • Outbound: DialogBot system calls a list of phone numbers from your campaign.
2

Speech recognition

The AI listens to what the person says and transcribes it into text in real time.
3

LLM decision

DialogBot passes the transcription through a large language model (LLM) together with your configured system prompt. The model crafts a reply or decides the next action — such as transferring the call or booking an appointment.
4

Voice output

The AI’s response is converted to natural-sounding speech via the voice engine (powered by ElevenLabs or Cartesia voices) and played back to the caller.
5

Result and actions

  • Inbound: The AI answers questions, verifies information, takes a message, or routes the caller.
  • Outbound: The AI pitches a product, qualifies a lead, or transfers an interested caller to a human agent.
6

Data and reporting

After the call, you receive transcripts, call recordings, call outcomes (such as “no answer” or “completed”), and the ability to trigger automated follow-up actions.

​
Inbound vs. outbound

  • A customer dials a number you rented or integrated with your existing phone system.
  • The AI answers, greets the caller, and handles queries or routes the caller if needed.
  • Useful for customer support, appointment booking, FAQ handling, and more.

​
Next steps

  • Learn the key terminology used throughout the platform.
  • Head to core concepts to understand the building blocks before configuring your first assistant.