# Cremona Image Editor

> The complete Cremona Image Editor documentation from https://docs.roscommon.systems/cremona/, with every page in one file.

## Cremona Image Editor

Source: https://docs.roscommon.systems/cremona/

Cremona is an AI image-editing assistant built for blind and low-vision users. Where traditional editors assume you can see the canvas, Cremona works entirely through conversation: you upload an image, Cremona describes what's in it, and you make changes by saying what you want. Its tagline puts it plainly — **image editing for the blind**.

Cremona runs in your web browser at [cremona.roscommon.systems](https://cremona.roscommon.systems/) — there is nothing to install. You control it with your voice, and it speaks back using one of over 20 natural voices across a range of languages and accents.

### Who Cremona is for

Cremona is designed for anyone who cannot rely on a visual editing interface:

- **Blind and low-vision users** who want to crop, adjust, and refine images independently
- **Screen reader users** who find conventional editors (with their toolbars, layers, and drag handles) impractical
- **Anyone** who would rather describe an edit than hunt for the right menu

No design experience is needed. If you can describe what you want, Cremona can help you do it.

### How Cremona works

1. **Sign in** with your Google account at [cremona.roscommon.systems](https://cremona.roscommon.systems/).
2. **Upload an image** — Cremona describes it so you know exactly what you're working with.
3. **Edit by voice** — say what you'd like changed, and Cremona applies it and tells you the result.

See the [Getting Started guide](https://docs.roscommon.systems/cremona/getting-started/) for a step-by-step walkthrough.

### Cremona and LIMA

Cremona and [LIMA Desktop](https://docs.roscommon.systems/lima/) are both from Roscommon Systems and share the same goal: making the screen usable through natural, AI-powered conversation.

| | LIMA | Cremona |
|---|---|---|
| What it does | Reads and controls your whole screen | Edits a single image in depth |
| Where it runs | Windows desktop app | Web browser — nothing to install |
| Best for | Day-to-day computer use | Focused work on photos and images |

The two are complements: use LIMA to navigate your computer, and Cremona when you want to work closely with an image.

### Next steps

- [Getting Started](https://docs.roscommon.systems/cremona/getting-started/) — sign in and make your first edit
- [Features](https://docs.roscommon.systems/cremona/features/) — what Cremona can do
- [Voice Commands](https://docs.roscommon.systems/cremona/voice-commands/) — the kinds of things you can say

### Using these docs with an AI assistant

All of the Cremona documentation is also available as a single text file: [Cremona documentation for AI assistants](https://docs.roscommon.systems/cremona/llms-full.txt). Give this link to an AI assistant and it can read every page at once, rather than one page at a time.

## Getting Started

Source: https://docs.roscommon.systems/cremona/getting-started/

Cremona runs entirely in your web browser — there is nothing to download or install. This guide takes you from signing in to your first edit.

### What you need

| Requirement | Details |
| --- | --- |
| A web browser | Any modern browser (Chrome, Edge, Firefox, Safari) |
| A Google account | Used to sign in — there is no separate password to create |
| An internet connection | Required — Cremona's descriptions and edits are powered by AI |
| Speakers or headphones | Cremona speaks its responses aloud |
| A microphone | Recommended, for controlling Cremona by voice |
| An image to edit | JPEG, PNG, or WebP, up to 10 MB |

### Step 1 — Open Cremona

Go to [cremona.roscommon.systems](https://cremona.roscommon.systems/) in your browser.

### Step 2 — Choose a voice

Before you sign in, you can pick the voice Cremona uses to speak to you from the voice menu. There are over 20 voices across several languages and accents — choose whichever you find clearest and most comfortable. You can change it later at any time.

### Step 3 — Sign in with Google

Select **Continue with Google** and choose your account. Cremona uses your Google account to sign you in, so there's no separate password to remember.

!!! tip "Use a consistent email"
    Sign in with the same Google account each time so your preferences and history stay with you. If you have purchased a Cremona plan, use the same email address you bought it with.

### Step 4 — Upload an image

Select **Upload Image** and choose a file from your device. Cremona accepts:

- **JPEG**, **PNG**, or **WebP** files
- Up to **10 MB** in size

Once the image loads, Cremona describes what it sees, so you know exactly what you're starting with before making any changes.

### Step 5 — Make your first edit

Ask Cremona to make a change by describing what you want in plain language. For example:

> *"Crop this to a square."*

Cremona applies the edit and tells you what it did. If it isn't quite right, just say so:

> *"Actually, make it a bit tighter around the face."*

You can keep refining through conversation until you're happy with the result.

!!! note "Voice and typing"
    You can speak to Cremona or type your requests — use whichever suits you. Everything Cremona can do is available either way.

### Next steps

- [Features](https://docs.roscommon.systems/cremona/features/) — the full picture of what Cremona can do
- [Voice Commands](https://docs.roscommon.systems/cremona/voice-commands/) — example phrases to try
- [Settings](https://docs.roscommon.systems/cremona/settings/) — change your voice and preferences

## Features

Source: https://docs.roscommon.systems/cremona/features/

Cremona brings image editing within reach for people who can't rely on a visual interface. Everything happens through conversation — you describe what you want, and Cremona does it and tells you the result.

### Conversational Image Description

When you upload an image, Cremona describes it in natural language — the subjects, the composition, the colours, and any text it can read. This gives you a clear mental picture of what you're working with before you change anything, and you can always ask for more detail:

> *"Describe the background in more detail."*

> *"Is there any text in this image? What does it say?"*

### Voice-Controlled Editing

You edit by describing the change you want, rather than finding a tool or menu. Cremona interprets your request, applies it, and confirms what it did — then you can keep refining through conversation.

Typical edits you can ask for include:

- **Cropping and framing** — *"Crop this to a square"*, *"Cut out everything except the dog"*
- **Rotating and straightening** — *"Rotate it 90 degrees to the right"*
- **Resizing** — *"Make this smaller for email"*
- **Colour and light** — *"Make it brighter"*, *"Turn this black and white"*
- **Retouching** — *"Remove the person in the background"*

After each change, Cremona tells you what it did so you always know the current state of the image, without needing to see it.

### Natural Voice Feedback

Cremona speaks its descriptions and confirmations aloud using natural, human-sounding voices. You can choose from **over 20 voices** spanning multiple languages and accents, so you can pick one that's clear and comfortable for you. See [Settings](https://docs.roscommon.systems/cremona/settings/) for the full list and how to change it.

### Dual Input — Voice or Typing

Every feature works whether you speak to Cremona or type your requests. Use your voice for a hands-free experience, or type when you're somewhere quiet or prefer the keyboard — you can switch freely at any time.

### Runs in Your Browser

Cremona is a web app — there's nothing to install or update. Sign in with Google from any computer and your experience is the same. See [Getting Started](https://docs.roscommon.systems/cremona/getting-started/) to sign in.

### Supported Image Formats

| Format | Supported |
| --- | --- |
| JPEG (`.jpg`, `.jpeg`) | Yes |
| PNG (`.png`) | Yes |
| WebP (`.webp`) | Yes |
| Maximum file size | 10 MB |

### Next steps

- [Voice Commands](https://docs.roscommon.systems/cremona/voice-commands/) — example phrases to try
- [Settings](https://docs.roscommon.systems/cremona/settings/) — personalise your voice and preferences
- [Support](https://docs.roscommon.systems/cremona/support/) — troubleshooting and contact details

## Voice Commands

Source: https://docs.roscommon.systems/cremona/voice-commands/

Cremona understands natural language — you don't need to memorise exact commands. The examples below show the kinds of things you can ask; any phrasing that communicates the same intent will work. You can speak these or type them.

### Understanding your image

| What you can ask Cremona to do | Example |
| --- | --- |
| Describe the whole image | *"What's in this picture?"* |
| Describe part of it in detail | *"Tell me more about the background"* |
| Read any text in the image | *"Is there any text? Read it to me"* |
| Ask about colours | *"What colours are in this image?"* |

### Cropping and framing

| What you can ask Cremona to do | Example |
| --- | --- |
| Crop to a shape | *"Crop this to a square"* |
| Crop to a subject | *"Cut out everything except the dog"* |
| Zoom in tighter | *"Make it tighter around the face"* |

### Adjusting the image

| What you can ask Cremona to do | Example |
| --- | --- |
| Rotate or straighten | *"Rotate it 90 degrees to the right"* |
| Resize | *"Make this smaller for email"* |
| Brightness and contrast | *"Make it brighter"* |
| Colour effects | *"Turn this black and white"* |

### Retouching

| What you can ask Cremona to do | Example |
| --- | --- |
| Remove something | *"Remove the person in the background"* |
| Clean up | *"Get rid of the spot in the corner"* |

### Getting the result

| What you can ask Cremona to do | Example |
| --- | --- |
| Confirm the current state | *"What does the image look like now?"* |
| Undo a change | *"Undo that"* |
| Save or download | *"Save this image"* |

### Adjusting the voice

| What you can ask Cremona to do | Example |
| --- | --- |
| Change voice | *"Switch to a different voice"* |
| Repeat the last response | *"Say that again"* |

!!! tip
    If an edit isn't quite what you wanted, just describe the correction — *"a little less"*, *"more to the left"*, *"undo that"*. Editing in Cremona is a conversation, not a one-shot command.

### Next steps

- [Features](https://docs.roscommon.systems/cremona/features/) — the full picture of what Cremona can do
- [Settings](https://docs.roscommon.systems/cremona/settings/) — change your voice and preferences

## Settings

Source: https://docs.roscommon.systems/cremona/settings/

Cremona is designed to fit how you like to work. The most important choice is the voice it uses to speak to you.

### Choosing a voice

Cremona speaks its descriptions and confirmations aloud, and you can choose from **over 20 voices** across several languages and accents. The voice menu is available when you open Cremona, and you can change your choice at any time — just pick the one that's clearest and most comfortable for you.

!!! info "About this list"
    The voices below are examples of what's on offer. The definitive, up-to-date list is always the voice menu in the app.

A selection of the available voices:

| Language / accent | Example voices |
| --- | --- |
| English | Ivy, James, Tyler, Winter, Bella, David, Kyle, Helen, Martha, River, Emma, Victor, Eleanor |
| Spanish | Mateo, Diego, Lucia |
| Italian | Giulia, Luca |
| French | Pierre |
| Hindi | Arjun |
| Russian | Dmitri |

!!! tip
    Try a few voices before settling on one. The right voice makes a real difference when you're listening to detailed image descriptions.

### Changing your voice

You can switch voices in the app's voice menu, or simply ask Cremona:

> *"Switch to a different voice."*

### Your account

Cremona signs you in with your Google account — there is no separate password to manage. To use a different account, sign out using the link in the app and sign back in with **Continue with Google**.

### Next steps

- [Features](https://docs.roscommon.systems/cremona/features/) — what Cremona can do
- [Support](https://docs.roscommon.systems/cremona/support/) — troubleshooting and contact details

## Support

Source: https://docs.roscommon.systems/cremona/support/

### Troubleshooting

#### I can't sign in

Cremona signs you in with your Google account — there is no separate password to enter.

1. Check your internet connection — signing in needs a working internet connection to confirm your account.
2. Select **Continue with Google** and make sure you complete the sign-in in the window that appears.
3. If you have a Cremona plan, sign in with the **same email address you used to purchase it**.
4. If the problem persists, try a different browser, or contact support with the email address you're signing in with.

#### My image won't upload

- Check the **format**: Cremona accepts JPEG, PNG, and WebP files.
- Check the **size**: files must be **10 MB or smaller**. If yours is larger, resize or re-export it first.
- Try a different browser if the upload button doesn't respond.

#### I can't hear Cremona

1. Check that your speakers or headphones are connected and the volume is up.
2. Make sure the Cremona tab isn't muted. Open the tab's context menu: if it offers an option such as **Unmute site** or **Unmute tab**, the tab is muted, so choose that option to turn the sound back on.
3. Ask Cremona to repeat itself — *"Say that again"* — to confirm audio is working.

#### Cremona isn't responding to my voice

1. Check that your microphone is connected and that your browser has permission to use it. When your browser asks whether Cremona can use your microphone, allow it. If you turned it down earlier, open your browser's site settings for cremona.roscommon.systems, allow the microphone, then reload the page.
2. Speak clearly and at a normal pace.
3. If voice input still doesn't work, you can **type** your requests instead — every feature works either way.

#### The description or edit isn't what I expected

- Cremona understands natural language, so try **rephrasing** — there's no single correct wording.
- Describe the correction rather than starting over: *"a little less"*, *"more to the left"*, *"undo that"*.
- Ask Cremona to describe the current state — *"What does the image look like now?"* — to check where things stand.

---

### Contact Support

If you can't resolve your issue with the steps above, our support team is here to help.

| Channel | Details |
| --- | --- |
| **Email** | hello@roscommonsystems.com |

!!! tip
    When contacting support, include the email address you sign in with, the browser you're using, and a brief description of what you were doing when the issue occurred. This helps us resolve your issue faster.