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

# Mazer Live (voice)

> Talk to Mazer in real time, hands-free, in one of eleven voices, and show it what you're looking at with Live Lens.

**Mazer Live** is a real-time voice conversation with Mazer. You talk, it answers out loud, and you can interrupt it like a person. It's on every plan, in the web app and the Android app.

## Start a conversation

**1. Open the chat**

Open [tripmazer.com/app](https://tripmazer.com/app), or the home-screen or Android app.

**2. Start Mazer Live**

Tap the **Mazer Live** button beside the message box (its label reads "Start Mazer Live — real-time voice conversation").

**3. Allow the microphone**

The first time, your browser asks for microphone access. Allow it.

**4. Talk**

Speak normally. To stop, tap the end button or close voice mode.

> [!TIP]
>
> Mazer Live is a full conversation. For a quick spoken question in the normal chat, use the **voice input** microphone instead, or send a voice note on WhatsApp or Telegram.

## What it can do

Mazer Live can do everything typed chat does. It reads the web, your memory, past chats, trips, bookings and connected apps. It can also change trips, itineraries, expenses, reminders, packing and connected apps.

Before any change, Mazer reads it back and waits for a clear **yes**. Payment still happens in the web checkout. Anything you tell Mazer in Live is remembered, just as in chat.

## Live Lens

**Live Lens** lets Mazer see through your camera while you talk: point it at a menu, a sign, a station board or a building and ask about it. Start it from the Live screen, and allow camera access when asked. If your browser doesn't support a live camera stream, take a snapshot instead.

## Choose a voice

Mazer has eleven voices, each with a real accent. Pick one in **Settings → AI Preferences** or from the voice picker:

| Voice | Accent |
| - | - |
| Kore | Neutral (the default) |
| Iapetus | Indian English, and Hinglish |
| Algenib | Manchester |
| Leda | Dublin |
| Aoede | Sydney |
| Pulcherrima | New Zealand |
| Umbriel | American Southern |
| Sadachbia | New York |
| Algieba | African American English |
| Zubenelgenubi | Egyptian English |
| Autonoe | Kenyan English |

Each voice uses a light touch of its everyday vocabulary, and drops it wherever precision matters: amounts, dates, booking references, cancellation terms and anything urgent. Mazer never imitates a real person.

Voice notes and read-aloud replies can carry natural reactions, such as a laugh, a sigh or a pause, usually one at most. That's in English for now. Booking references, codes and phone numbers are read slowly, and delays, cancellations and emergencies are read gently.

The same voice is used for voice notes on WhatsApp and Telegram. To have written replies carry it too, turn on **Write in this voice too** in **Settings → AI Preferences**.

## Steer it mid-conversation

You don't need settings to change how Mazer talks. Just say it:

* "Slower, please." or "You can speed up."
* "Quieter." or "Keep it short."
* "Switch to the Dublin accent."

The change applies straight away.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.