# Voices, Greetings and Pronunciation

Everything on the **Voice** tab that governs how the agent *sounds*: the voice
itself, the first thing it says, tone direction, pronunciation, and the room
it appears to be sitting in. Turn-taking and the silence timers are on the
same tab but are a subject of their own — see
[Turn-Taking, Interruptions and Dead Air](/help/voice-agents/turn-taking-and-timing).

## What you'll need

- The **`agents:voice:manage`** permission.
- A saved voice agent — see [Create a Voice Agent](/help/voice-agents/creating-campaigns).

## Choose a voice

1. Open the agent and click the **Voice** tab.
2. At the top, type a phrase into **Preview Text** — or click **Use agent
   greeting** to load your own greeting, which is the better test. **Reset**
   restores the default phrase. The limit is 500 characters.
3. Below it, filter the grid with the search box ("Search voices by name,
   language, or style…") or the category pills: **All · OpenAI · English ·
   Español · Français · हिंदी · ਪੰਜਾਬੀ · Emotive** (plus **Gemini** when
   Gemini voices are enabled for your workspace), each showing how many
   voices it holds.
4. Press the play button on any card to hear that voice read your preview
   text.
5. Click a card to select it. The selected voice is shown beside the **Voice**
   label at the top of the picker, with its own play button.

52 voices ship today in two families, and a third family of 30 appears when
Gemini voices are enabled for your workspace. Every card shows the voice's
language, gender and style on one line and, in small type beneath, the
**provider and model** that speaks it:

| Family | Count | Small label | Notes |
|---|---|---|---|
| **Cartesia** | 42 | `Cartesia · Sonic 3.6` | English (incl. `EN-US`, `EN-GB`), Spanish, French, Hindi and Punjabi voices, plus ten tuned for emotional range (`Emotive`) |
| **OpenAI** | 10 | `OpenAI · gpt-4o-mini-tts` | Alloy, Ash, Ballad, Coral, Echo, Fable, Onyx, Nova, Sage, Shimmer — take a written **Speaking style** |
| **Gemini** (beta) | 30 | `Google · Gemini 3.8 Flash-Lite TTS` | Google's prebuilt voices (Kore, Puck, Achird, Sulafat, …), each described in a word (Firm, Upbeat, Friendly, Warm). Every one speaks every language, and they take a **Speaking style**. A super admin enables them per workspace while they are in beta; until then they are not listed, and an agent cannot be saved with one |

**Your voices** appears first when your workspace has custom voices: a person
on your team cloned with their consent, or a voice designed from a
description. Each is marked **Yours**. Calls in one add $0.06 a minute, and an
agent on one goes live only when its opening says it is an AI and that the
call may be recorded — see [Custom Voices](/help/voice-agents/custom-voices).

If the agent already holds a voice id that is not in the grid, the picker says
so underneath: *"Custom voice ID in use: …"*. It still works; it simply is not
one of the listed options. When the id is one of your custom voices that can no
longer speak (deleted, or custom voices switched off), the picker names the
voice and says calls speak a standard voice instead.

**The choice decides which speech engine runs the call** — the small label
on the card names it. That is why the next setting only appears for some
voices.

## Steer the tone

**Conversation Behavior → Speaking style** is a short line on how the voice
delivers every sentence — "warm and unhurried", "brisk and factual". The
preview buttons use it, so you hear the style before you save.

**It appears only for OpenAI and Gemini voices**, the two engines that take a
written style. Keep it short: a long description makes a Gemini voice drift.
Cartesia voices take their tone from the words themselves, so the field is
hidden for them; if you switch an agent with a style to a Cartesia voice, the
text is kept and applies again if you switch back. To change a Cartesia
voice's feel, change the voice.

Tone that belongs in the *instructions* instead: how long turns should be,
whether to use the caller's first name, how formal to be. Those are
behavioural, and they work on every voice.

## Decide who speaks first

**Conversation Behavior → Who Speaks First?** has two values:

- **User Initiates** — the default. The agent stays silent until the caller
  speaks. On an inbound line this means the caller hears the line connect and
  then nothing, so most inbound agents should not be left here.
- **AI Initiates** — the agent opens the call. This is what plays your
  **Static Greeting**.

The greeting itself is on **Setup → Instructions & Greeting → Static
Greeting**. It is spoken **verbatim**, so it is the one line of the call you
control exactly. `{{first_name}}`, `{{last_name}}` and `{{caller_number}}`
resolve in it; an unknown caller resolves them to empty strings, so
`Hi {{first_name}}, thanks for calling` becomes `Hi , thanks for calling` —
write it as `Hi, thanks for calling` and let the instructions handle names.

Outbound calls behave differently on purpose: the agent **waits for the callee
to say something** (up to 30 seconds) before it plays its opener, and plays it
without allowing an interruption. Picking up a call and being talked over is
the thing that makes an outbound robot obvious.

> **Voice delivery tags in the greeting are deliberate.** The static greeting
> is the one place emotion markup may be authored inline, and it appears as
> written in the stored transcript. That is expected. Anywhere *else* in a
> call, the platform rules forbid the agent emitting those tags and they are
> stripped — the browser test panel flags such a turn with a small **tags**
> marker so you can report it.

## Voicemail handling

**Conversation Behavior → Voicemail** decides what the agent does when an
outbound call reaches an answering machine: leave a message **in its own
words** (default — guided by your **Voicemail instructions**), say an **exact
message**, or **hang up without a message**. **Detection hints** help the
platform recognise your customers' machines. Details and the full table are
in [Placing Outbound Calls](/help/voice-agents/campaign-usage#voicemail).

## Teach it your words

> **Paused on calls (October 2026).** Agents currently say every word as
> written while we test which entries the latest voices still need. Your
> packs and workspace entries are kept and apply again when pronunciation is
> switched back on.

**Pronunciation Packs** (below the voice picker) rewrite words on their way to
the speech engine, so your product and place names come out right instead of
phonetically plausible. Packs come in four groups:

| Group | Behaviour |
|---|---|
| **Always-On (Platform Core)** | Applied to every agent; not switchable |
| **Regional Packs** | Opt in per agent — US, UK, Canadian English and French, French, Spanish and Mexican Spanish |
| **Vertical Packs** | Opt in per agent — automotive, finance, healthcare, home services, hospitality, legal, real estate |
| **Tenant Custom** | Your workspace's own entries; always applied |

Tick the packs this agent should carry. Each row shows the pack name, what it
covers and an entry count.

A word still coming out wrong after the right pack is on usually needs a
workspace entry rather than a pack — that is what **Tenant Custom** is for.

## Background audio

**Background Audio** (default **on**) plays quiet office ambience and keyboard
typing while the agent is thinking. It exists because dead silence during a
pause reads as a dropped call, and a caller who thinks the line dropped hangs
up.

It only plays on telephony calls. Browser calls never get it, whatever this
switch says.

Turn it off for agents where the illusion is wrong — an explicitly automated
line, or a clinical or legal context where ambience is out of place.

## Live-audio models (Gemini Live)

If a super admin has enabled **Gemini Live voice agents (beta)** for your
workspace, the **AI Model** list on the Setup tab also offers **Gemini 3.8
Live** and its **extended thinking** variant. On one of these a single Google
model listens and speaks: there is no separate speech-to-text or
text-to-speech step, and the Voice tab changes to match.

- **Voice** — the picker shows Gemini voices only; the live call speaks with
  the one you pick (any other voice would run as Puck). Each speaks every
  language.
- **Speaking style**, **Fillers** and **Pronunciation packs** do not apply;
  what you saved is kept for the normal pipeline. Spell a hard name out in the
  instructions instead.
- **Static Greeting**, the **exact voicemail message** and the **Exact line**
  options are greyed out: with a live-audio model the agent speaks from your
  instructions, so put the exact wording there.
- **Fallback voice** — if the live session cannot connect or drops mid-call,
  the call carries on in the normal pipeline, conversation so far included, in
  this Google voice (the agent's own by default; Achird, Sulafat and Iapetus
  are recommended for a smooth hand-over). The call page says when that
  happened and why.
- **Languages the agent may speak** — a checklist of the languages Google's
  Live models support. The agent answers in the caller's language when it is
  ticked, otherwise in the first ticked one; nothing ticked means the Default
  Language alone. See [Multilingual Calls](/help/voice-agents/languages).

## Troubleshooting

- **The Voice tab shows a "!" badge.** No voice is selected. The agent cannot
  be promoted usefully without one.
- **A preview plays nothing.** Check your browser's audio output and that
  **Preview Text** is not empty.
- **There is no Speaking style field.** The selected voice is a Cartesia voice;
  only OpenAI and Gemini voices take a written style.
- **No Gemini voices in the picker.** They are a beta a super admin enables per
  workspace. An agent that already uses one keeps it, and the picker says so.
- **Callers hear silence when they connect.** **Who Speaks First?** is on
  **User Initiates**. Switch to **AI Initiates** and set a **Static
  Greeting**.
- **The greeting says "Hi ,".** A `{{first_name}}` placeholder resolved empty
  for an unknown caller. Remove it from the greeting.
- **Emotion tags are readable in a transcript.** In the static greeting, that
  is by design. In any other assistant turn it is a fault worth reporting.

## Next steps

- [Turn-Taking, Interruptions and Dead Air](/help/voice-agents/turn-taking-and-timing)
- [Multilingual Calls and Language Detection](/help/voice-agents/languages)
- [Instructions, Goals and Scheduled Prompts](/help/voice-agents/instructions-and-scheduling)
- [Testing, Drafts and Going Live](/help/voice-agents/testing-and-promotion)