> ## Documentation Index
> Fetch the complete documentation index at: https://docs.hiveku.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Set Up the AI Voice Receptionist

> Configure the AI to answer your phone, capture leads, and book meetings

This guide gets the Hiveku AI Voice Agent answering your business phone — greeting callers, understanding intents, capturing details, booking meetings, and transferring to humans when appropriate. Plan for 30-60 minutes including testing and refinement.

## Before you start

Have ready:

* At least one DID configured. See [Buy and Configure a Phone Number](/how-tos/buy-phone-number)
* A clear greeting script (or willingness to draft one in the builder)
* Decisions about which intents you want handled and where they should route
* Optional but useful: FAQ content, pricing, or product specs to upload as the agent's knowledge base

## Why the AI agent vs an IVR

| Use IVR                                 | Use AI agent                             |
| --------------------------------------- | ---------------------------------------- |
| 2-3 simple, well-known options          | Open-ended caller intent                 |
| You already have a recorded brand voice | You want 24/7 conversational handling    |
| Compliance forbids LLM-driven outputs   | LLM-quality answers to factual questions |
| Calls always need a human               | Many calls can be fully self-served      |

The two aren't exclusive — many businesses use the AI agent during business hours and a simpler IVR after-hours, or vice versa.

## Step 1: Pick a voice persona

<Steps>
  <Step title="Open the AI Agent tab">
    `/dashboard/communications/` > **AI Agent**.
  </Step>

  <Step title="Browse personas">
    Multiple voice options — male/female, US/UK/AU accents, neutral or warm tone. Each has an inline preview button.
  </Step>

  <Step title="Pick one">
    Match your brand. Casual D2C → warmer, friendlier. Enterprise B2B → measured, professional. You can change later.
  </Step>
</Steps>

## Step 2: Write the greeting

The greeting is the first thing callers hear. Short and friendly, with an open question that invites the caller to say what they need.

```
"Hi, thanks for calling Acme. How can I help you today?"
```

Avoid menu-style greetings ("press 1 for sales") — the AI is conversational, not a phone tree.

<Steps>
  <Step title="Write the greeting in the builder">
    Type your script. Preview the audio.
  </Step>

  <Step title="Iterate">
    Listen 3-5 times. Adjust word choice and timing until it sounds natural.
  </Step>
</Steps>

## Step 3: Define intents

Intents are the categories of caller need the agent should recognize. Most businesses have 3-6 intents.

```
sales
support
billing
partnerships
careers
general (catch-all)
```

For each intent, decide what the agent does:

<Tabs>
  <Tab title="Capture and transfer">
    Get the caller's name, company, email, and reason. Transfer to a ring group or queue with a verbal hand-off summary. Use for: sales, support.
  </Tab>

  <Tab title="Book a meeting">
    Connect to your [Calendar](/calendar/overview). Offer slots and book directly on the call. Use for: sales discovery, demos.
  </Tab>

  <Tab title="Answer from KB">
    Look up the answer in your knowledge base and reply inline. Use for: pricing, hours, FAQ.
  </Tab>

  <Tab title="Take a message">
    Capture details and email/Slack a summary to the right team. Use for: after-hours, low-priority intents.
  </Tab>

  <Tab title="Always transfer">
    Handoff immediately. Use for: emergencies, billing disputes, anything sensitive.
  </Tab>
</Tabs>

## Step 4: Build the knowledge base

The KB is what the agent uses to answer factual questions inline. Without it, the agent will say "let me transfer you" for any specific question.

<Steps>
  <Step title="Open the KB tab in the AI Agent settings">
    Knowledge Base section.
  </Step>

  <Step title="Add sources">
    Three input modes:

    * **URL** — Hiveku crawls the page (e.g., your pricing page, FAQ, support docs)
    * **Upload** — PDFs, Markdown, plain text
    * **Direct entry** — Paste content directly
  </Step>

  <Step title="Tag entries">
    Tag entries to specific intents. "Pricing FAQ" tagged with `sales` so the agent only references it on sales calls.
  </Step>

  <Step title="Test KB lookups">
    Use the inline tester: "What's the price of the Pro plan?" The agent shows which KB entry it would cite.
  </Step>
</Steps>

<Tip>
  Start with your most-asked questions — the top 10 questions you get on calls today. That alone will deflect 60-80% of calls from human handoff.
</Tip>

## Step 5: Set escalation rules

Decide when the agent should transfer to a human.

<CardGroup cols={2}>
  <Card title="Caller asks for a person" icon="user">
    "Can I speak to someone?" — always transfer.
  </Card>

  <Card title="Frustration detected" icon="face-frown">
    Profanity, raised tone, repeated complaints — transfer.
  </Card>

  <Card title="Stuck on a step" icon="circle-question">
    3 reprompts on the same input (email, name) — transfer.
  </Card>

  <Card title="Out-of-scope intent" icon="signpost">
    Caller's need doesn't match any handler — transfer.
  </Card>

  <Card title="Always transfer intent" icon="forward">
    Specific intents (e.g., billing dispute) always go to a human.
  </Card>

  <Card title="Configurable timeout" icon="hourglass">
    If the agent takes longer than N minutes total, transfer (rare safety net).
  </Card>
</CardGroup>

Pick a transfer destination — ring group, queue, or specific extension. The receiving agent gets a brief verbal summary of what the AI gathered: "I have Sarah from Acme on the line, she's interested in the Pro plan, here she is."

## Step 6: Connect calendar (for in-call booking)

If you want the agent to book meetings during a call, connect [Calendar](/calendar/overview) and pick which meeting type it can book.

<Steps>
  <Step title="In AI Agent settings, find 'Booking'">
    Toggle on.
  </Step>

  <Step title="Pick a meeting type">
    Usually the most common inbound meeting — "Demo (30 min)" or "Intro Call (15 min)".
  </Step>

  <Step title="Set host routing">
    Single host (always books with one person), team round-robin, or specific-host based on intent.
  </Step>
</Steps>

## Step 7: Point a DID at the agent

<Steps>
  <Step title="Open the DID's detail page">
    `/dashboard/communications/` > **Numbers** > pick the DID.
  </Step>

  <Step title="Set inbound destination to AI Agent">
    Replace whatever was there.
  </Step>

  <Step title="Set time conditions if needed">
    Common pattern: AI agent during business hours, simpler "leave a message" IVR after-hours. Or AI agent 24/7.
  </Step>
</Steps>

## Step 8: Test thoroughly

This is where most of the value of setup time is spent. The first time you call, you'll hear three things you want to fix.

<Steps>
  <Step title="Call from your own phone">
    Run through 3-5 scenarios:

    * "I'm interested in your product" — sales path
    * "I have a billing question" — billing path
    * "What are your hours?" — KB path
    * "Can I talk to a person?" — escape hatch
    * Long silence — does it reprompt?
  </Step>

  <Step title="Listen to transcripts">
    Open each call in `/dashboard/communications/calls/`. Read the transcript. Note where the agent answered well and where it stumbled.
  </Step>

  <Step title="Refine">
    For factual misses → improve KB entry phrasing. For tonal issues → change persona or adjust prompt instructions. For routing issues → update intent definitions.
  </Step>

  <Step title="Test edge cases">
    Background noise, slow connections, unusual accents, very fast speech, very slow speech.
  </Step>
</Steps>

## Step 9: Soft launch

Before pointing your main number at the agent, soft-launch on a secondary DID.

<Steps>
  <Step title="Buy a second DID">
    Use it for outbound or one specific marketing campaign so you can monitor inbound carefully.
  </Step>

  <Step title="Run for 1-2 weeks">
    Listen to every call, refine the KB and prompts.
  </Step>

  <Step title="Monitor metrics">
    Containment rate (calls handled fully by AI), transfer rate, customer satisfaction (NPS via post-call SMS).
  </Step>

  <Step title="Cut over your main number">
    Once metrics look good — typically 70%+ containment with positive NPS — point your main DID at the agent.
  </Step>
</Steps>

## Ongoing refinement

Plan to spend 30-60 minutes/week reviewing flagged calls and refining the agent for the first month. After that, weekly review of outliers is enough.

| Signal                              | Action                                          |
| ----------------------------------- | ----------------------------------------------- |
| KB miss (agent said "I don't know") | Add a KB entry                                  |
| Wrong intent classification         | Improve intent description, add example phrases |
| Bad transfer summary                | Refine the handoff prompt template              |
| Caller frustration                  | Re-listen, identify trigger, adjust             |

## Cost monitoring

The AI agent uses LLM tokens and TTS/STT. Usage and cost are visible in the **Billing** tab. A typical 3-minute conversation costs roughly the same as a few sales emails. See [Monitor Usage](/how-tos/monitor-usage).

## Troubleshooting

<AccordionGroup>
  <Accordion title="Agent sounds robotic">
    Try a different persona — voice quality varies meaningfully. Also keep the agent's response prompts short; long replies sound more synthetic.
  </Accordion>

  <Accordion title="Caller can't get to a human">
    Make sure escalation rules include "if caller asks for a person" and that the transfer destination is configured. Test the escape hatch yourself.
  </Accordion>

  <Accordion title="Agent gives wrong factual answers">
    Check the call transcript for the cited KB source. Update the source, or add a more direct phrasing. The KB is what grounds factual answers.
  </Accordion>

  <Accordion title="Calls cut off mid-sentence">
    Voice-activity detection may be too aggressive in noisy environments. Lower the noise sensitivity in advanced settings.
  </Accordion>
</AccordionGroup>

## Next steps

<CardGroup cols={2}>
  <Card title="Voice agent reference" icon="circle-info" href="/communications/voice-ai-agent">
    Full feature reference.
  </Card>

  <Card title="Compliance" icon="shield-check" href="/communications/compliance">
    Recording controls and retention for AI-handled calls.
  </Card>
</CardGroup>