Configuring an agent¶

This page covers who the agent is — its persona, voice, and what it knows and asks. For how it runs the conversation, see Conversation design; for hand-off, see Call forwarding.

Persona¶

A persona is a ready-made package of voice + personality + language, shown with a photo and a short description (for example "Warm and professional British receptionist"). Choosing a persona sets a sensible default voice, greeting, and language in one click.

Open Change persona on the agent page to browse personas.
Personas are grouped by language — pick the language first, then the voice.
The persona's avatar follows the agent everywhere it appears (the agents list, the edit page), so the picture always matches the selected voice.

Some personas are restricted

A few personas (for example Hebrew-language voices) are only available to super-admin users.

Voice & language¶

Each persona maps to a specific voice from a speech provider. The agent speaks — and expects to hear — the persona's language. If a caller switches language mid-call, the agent keeps replying in its configured language unless told otherwise.

Greeting & opening question¶

The greeting is the first thing the agent says when it picks up, for example:

"Thank you for calling Elegant Homes, this is Grace. How can I help you today?"

A good greeting names the business, names the agent, and invites the caller to explain why they're calling. The caller's answer is scanned immediately for anything that already answers one of your intake questions, so the agent never re-asks for something the caller just told it.

Knowledge base¶

The knowledge base is free-text the agent can use to answer factual questions about the business — hours, services, pricing, location, process. If a caller asks something covered here, the agent answers first, then continues collecting any information it still needs.

Answer first, collect second

Callers' questions take priority. The agent will answer from the knowledge base before pushing on with intake — this feels natural and avoids interrogations.

Intake questions¶

Intake questions are the structured information you want from every caller — name, phone number, reason for calling, and so on. Each question has:

the spoken question the agent asks,
an internal field name used to store the answer (never read aloud), and
a type (text, phone, email, …) and whether it's required.

The agent asks one question at a time, skips anything the caller already provided, and validates answers as they're given (for example, reading an email address back phonetically to confirm spelling).

Reorder questions by dragging them; the order is the order they're asked.

Saving¶

Changes autosave as you edit. A status indicator shows "All changes saved" once a change has persisted.

Conversation design — Choose how an agent runs a call — simple intake, Questions & conditions (scenario), or a Decision tree.
Call forwarding & transfers — Set the conditions under which an agent hands a call to a human, and understand exactly how the transfer behaves.
Agents — What an agent is and the three things you configure — who it is, how it converses, and when it transfers.
Web calls & embeds — Let website visitors talk to an agent from a button — create an embed, set its origins, and install the snippet.
Getting started — From signing in to your first live call — the fastest path through Vocadesk.