> For the complete documentation index, see [llms.txt](https://evidano.gitbook.io/ailyze-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://evidano.gitbook.io/ailyze-docs/reference/settings/image-to-text.md).

# Image to Text Settings

Every control in the Image to Text wizard, project list, and page viewer, including the language and domain catalog, sample test, and charge dialog.

This page lists every control in the **Image to Text** workflow and says what each one changes: the three wizard steps, the charge dialog, the page list with its sidebar and bulk actions, and the page viewer. It is for anyone who has run [Convert Your First Image to Text](/ailyze-docs/getting-started/convert-your-first-image-to-text.md) and wants to know what a specific button does. The walkthrough with advice on choosing domains and testing prompts is [Convert Images & PDFs to Text](/ailyze-docs/prepare-research-data/convert-images-and-pdfs-to-text.md).

Pages beyond the free allowance are charged per page from your wallet on every plan, at a higher rate for the advanced model; see [Plans, Limits & Credits](/ailyze-docs/reference/plans-limits-and-credits.md). Accepted formats and caps are on [Supported File Types](/ailyze-docs/reference/supported-file-types.md). The numbers in each table match the red badges in the figure above it.

Last verified against the staging app on 14 October 2026.

## Step 1 of 3: Add files

From your account page, click **Image to Text** (or **View previous** for the list). The wizard's step bar reads **Add files**, **Language & domain**, and **Review and create**; a finished step's circle can be clicked to go back to it.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-7fc71998b5da5f4be2aaf7932cce390e78473a0b%2Fsettings-image-to-text-01-add-files.png?alt=media" alt="Image to Text Add files step with Previous Images, a project name, four source tabs and the drop zone"><figcaption><p>The drop zone lists the accepted formats and reminds you that images and PDFs cannot be mixed.</p></figcaption></figure>

| #   | Control                   | What it does                                                         | Notes                                                                                                                                                           |
| --- | ------------------------- | -------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| 1   | **Previous Images**       | Opens the Image to Text list.                                        |                                                                                                                                                                 |
| 2   | **Project name**          | Names the project. Blank uses today's date.                          | You can change it until you leave this step, except while an upload is running. With several PDFs, each PDF becomes its own project with the PDF name appended. |
| 3   | **Local Files**           | Drag images or PDFs onto the drop zone, or click **Browse files**.   |                                                                                                                                                                 |
| 4   | **Google Drive**          | **Open Google Drive** opens its picker.                              | Only the files you pick are shared with Evidano.                                                                                                                |
| 5–6 | **Dropbox**, **OneDrive** | Browse your cloud storage, tick files, and click **Import … files**. | A counter shows how many of the project's slots are left.                                                                                                       |
| 7   | **Browse files**          | Opens your computer's file picker.                                   |                                                                                                                                                                 |

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-1b4b3d1bd61a2021c8b93a61d7c69c1575cd29bc%2Fsettings-image-to-text-02-google-drive.png?alt=media" alt="Google Drive tab with the Open Google Drive button"><figcaption><p>Sign in to Google once; the picker then opens straight away.</p></figcaption></figure>

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-c7e33cf2b03d7e7a0e83bf33ab7a3dc96d202810%2Fsettings-image-to-text-03-files-added.png?alt=media" alt="Two uploaded field-note pages with preview and remove icons, a files-allowed counter, Back and Next Step"><figcaption><p>Files upload the moment they are added, and the project exists on the server from that point.</p></figcaption></figure>

| # | Control            | What it does                                                                               | Notes                                                                                                                             |
| - | ------------------ | ------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------- |
| 1 | **Preview file**   | Opens the page full screen (next figure).                                                  |                                                                                                                                   |
| 2 | **Remove file**    | Deletes the page from the project on the server.                                           |                                                                                                                                   |
| 3 | Files-allowed line | "Images only" plus how many more files fit, or "PDFs only · each becomes its own project". | A PDF row reads "Preparing pages…" and then its page count; every page counts toward the allowance and the price.                 |
| 4 | **Back**           | Leaves the wizard for the Image to Text page.                                              | Your uploads stay in the project.                                                                                                 |
| 5 | **Next Step**      | Moves to the language step.                                                                | Disabled until every upload finishes. "The server is still splitting these files into pages" means wait a moment and click again. |

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-d877922793682b1598883507f6aa3c03f0f54970%2Fsettings-image-to-text-04-preview.png?alt=media" alt="Full-screen preview of an uploaded field-note page with zoom buttons and file navigation"><figcaption><p>Use the arrow keys or Previous file and Next file to move through the pages.</p></figcaption></figure>

The full-screen preview has **Zoom out**, **Zoom in**, **Fit to view**, **Previous file**, **Next file**, an "N of M" counter, and **Close preview**. PDFs open in a plain viewer without zoom buttons.

## Step 2 of 3: Language & domain

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-f5057989acbf67958edc52de703c9a7ebc3e6c3b%2Fsettings-image-to-text-05-language-domain.png?alt=media" alt="Language and Domain columns with English and Scanned Documents selected, the added prompt text and the sample page bar"><figcaption><p>The box under the columns shows the exact wording the chosen domain adds to the transcription prompt.</p></figcaption></figure>

| # | Control                   | What it does                                                                                                                                   | Notes                                                                                                                         |
| - | ------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- |
| 1 | Search box                | Matches language names, domain names, and the wording of each domain's instructions.                                                           | Searching for "invoices" or "Kurrent" finds the right row. **Clear search** restores the full lists.                          |
| 2 | **Language** column       | The language of the writing. **Auto-detect** at the top detects it from each page; **Custom** at the end lets you type one that is not listed. | Changing the language clears the domain.                                                                                      |
| 3 | **Domain** column         | The kind of material, within that language. **Custom** lets you describe the pages yourself.                                                   | Most languages offer only **General**.                                                                                        |
| 4 | **ADDED TO YOUR PROMPT**  | The instructions the chosen domain contributes.                                                                                                | Read-only here; you edit the full prompt on the next step. **View guides** opens the reading guides when the domain has them. |
| 5 | **Test a different page** | Opens the sample picker. The bar names the **Sample page** the next step tests.                                                                | Shown when there is more than one page.                                                                                       |
| 6 | **Back to Previous Step** | Returns to **Add files**.                                                                                                                      |                                                                                                                               |
| 7 | **Next Step**             | Moves to the review step.                                                                                                                      | Needs a language and a domain, or **Auto-detect**, or a typed custom language.                                                |

The sample page is transcribed in the background as soon as the language and domain are set, so the next step is usually ready when you get there.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-c5c148f4cb1168cee101960a35f784654ef52a5c%2Fsettings-image-to-text-06-auto-detect.png?alt=media" alt="Auto-detect chosen as the language, with the Domain column reading No domain needed"><figcaption><p>Auto-detect suits mixed or unknown languages; the prompt then asks for the original language.</p></figcaption></figure>

With **Auto-detect** or a **Custom** language, the domain column reads "No domain needed". Projects set to Auto-detect show the language as "Original language" in the list and viewer.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-919ec127aa63c2a47159e628b02a06d474f87473%2Fsettings-image-to-text-07-custom-domain.png?alt=media" alt="Custom domain with the What should the model expect on these pages? box filled in"><figcaption><p>Describe the script, abbreviations, and layout, as you would brief a research assistant.</p></figcaption></figure>

The **Custom** domain's box, **What should the model expect on these pages?**, is added to the transcription prompt you review in the next step.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-a2346d896575c2c489045c6144f0058f6d3e0c3d%2Fsettings-image-to-text-08-guides.png?alt=media" alt="Reading guides dialog for an early modern English secretary hand with links to external references"><figcaption><p>Each guide is an external reference for reading that hand.</p></figcaption></figure>

The info icon on a domain (**View palaeography guides**) opens the reading guides for that hand. Domains without guides have no icon.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-e5dea15d919a5a92bc13e6ff55132d746f2fb095%2Fsettings-image-to-text-09-sample-picker.png?alt=media" alt="Sample page picker with two page thumbnails and Pick at random"><figcaption><p>Pick the hardest page as the sample: if the prompt copes with it, it copes with the rest.</p></figcaption></figure>

In the picker, click a page to select it or open it full screen to read it first. **Pick at random** chooses one for you; **Hide pages** closes the picker.

<details>

<summary>What the language and domain catalog contains</summary>

* **Auto-detect** and **Custom** sit at the two ends of the language list; most of the languages between them offer a single **General** domain that asks for a faithful transcription.
* English has five dated historical hands, from Old English manuscripts to Georgian administrative records, plus ten domains for modern material: **Handwritten Content**, **Forms and Questionnaires**, **Scanned Documents**, **Business and Financial Documents**, **Screenshots and Digital Captures**, **Educational Materials**, **Whiteboards and Workshop Materials**, **Signs, Labels, and Printed Materials**, **Official and Identity Documents**, and **Tables and Structured Content**.
* Latin, German, French, Spanish, Arabic, Ottoman Turkish, Italian, Dutch, Portuguese, Russian, Chinese (especially Classical Chinese), Japanese, Persian, Hebrew and Jewish documentary languages, Greek, Sanskrit, Polish, Czech and other Central European languages, and Swedish and Danish-Norwegian each offer five dated historical hands, most with reading guides.
* Each domain's instructions describe the letterforms, abbreviations, and layout features to expect, which is why the right pair matters more than anything you type into the prompt.

</details>

## Step 3 of 3: Review and create

The step opens with "Review the sample transcription": the page you picked, transcribed with the prompt shown. Sample runs are free and do not use your allowance, but only a handful can run each minute.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-c1c13c1145fe369a812ece73ce1a14265d3e378b%2Fsettings-image-to-text-10-review.png?alt=media" alt="Review step with the transcription prompt, the sample page beside its transcription, and the two verdict cards"><figcaption><p>The verdict cards decide which model runs on every page, so read the sample before choosing.</p></figcaption></figure>

| # | Control                                                                | What it does                                                                                    | Notes                                                                                                                    |
| - | ---------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------ |
| 1 | **Transcription prompt**                                               | The instruction the model follows: a fixed base plus your domain text. Editable.                | Editing it withdraws any approval, so test again. Once edited, going back and changing the domain no longer rewrites it. |
| 2 | **Change page**                                                        | Opens "Choose the page to test".                                                                | Shown when there is more than one page.                                                                                  |
| 3 | **Transcribe this page** / **Transcribe again**                        | Runs the sample with the current prompt.                                                        | A stop button cancels a run.                                                                                             |
| 4 | Image pane                                                             | The page, with **Zoom out**, **Zoom in**, and **Fit to view**.                                  | Drag the divider to resize the panes; double-click it to reset.                                                          |
| 5 | **Copy**                                                               | Copies the sample text.                                                                         | An **Advanced model** chip marks a sample produced by the advanced model.                                                |
| 6 | **Transcription looks good, proceed with this prompt for all N files** | Approves the prompt and starts the job on every page.                                           | Uses whichever model produced the sample on screen, so an advanced sample means an advanced run.                         |
| 7 | **Transcription looks bad**                                            | Offers **Edit prompt and try again** and **Retry with a more advanced AI model** (next figure). |                                                                                                                          |
| 8 | **Back to Previous Step**                                              | Returns to the language step.                                                                   |                                                                                                                          |

**Skip this sample transcription and transcribe all images** skips the test: the prompt as written is used for every page with the basic model, via **Start transcription**. **Start transcription** is shown only while no sample is on screen.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-2a4f8adabb8f9cd396a176c855e97e467730cb93%2Fsettings-image-to-text-11-looks-bad.png?alt=media" alt="Transcription looks bad card with Edit prompt and try again and Retry with a more advanced AI model"><figcaption><p>The advanced model reads faint, cursive, or damaged pages better, at a higher price per page.</p></figcaption></figure>

* **Edit prompt and try again** hides the cards and asks you to edit the prompt above, then **Transcribe again** to compare. **Keep this result** brings the cards back without a new run.
* **Retry with a more advanced AI model** runs the sample again with the advanced model. After an advanced sample it is greyed out: "This page was already transcribed with the more advanced model."

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-53a2ff6c34bdccea2461dce5ecff1638a8525734%2Fsettings-image-to-text-12-change-page.png?alt=media" alt="Choose the page to test dialog with page thumbnails and Cancel"><figcaption><p>Switch to a harder page to stress-test the prompt.</p></figcaption></figure>

{% hint style="warning" %}
**Costs money:** When a run goes past your free allowance, the **Confirm image transcription charge** dialog says how many pages the allowance covers, how many will use the basic or advanced model and at what price per page, and shows your **Current balance**, the **Transcription charge**, and the **Final balance**. **Start transcription** there confirms. With more than ten pages it also offers **Test with 10 images first**, which opens "Choose 10 images to test"; the other pages stay uploaded as **Not started**. If your balance is short, the dialog offers **Top up your wallet**. Accounts without a paid plan cannot start more pages than their remaining free allowance.
{% endhint %}

On success a message says how many files are queued and the **Image to Text** list opens. This workflow sends no email; the list and the viewer update themselves while pages are transcribing.

## The Image to Text list

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-1febc817c97b5439b3e2bbaac79a779a82e899db%2Fsettings-image-to-text-13-list.png?alt=media" alt="Image to Text list with the Projects sidebar, search bar with Search in, toolbar and page cards"><figcaption><p>The list shows pages, not projects; click a project in the sidebar to open it.</p></figcaption></figure>

| # | Control                       | What it does                                                                                                                                                                                                                                            |
| - | ----------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| 1 | **Projects (N)** sidebar      | Every project, with its file count and whether any files are not started yet. Click a project to open it. Hover it for **Rename project** and **Download project spreadsheet**. With more than five projects, **Search projects…** filters the sidebar. |
| 2 | **Hide projects panel**       | Collapses the sidebar (**Show projects panel** brings it back).                                                                                                                                                                                         |
| 3 | Search box                    | Finds pages by name, or by words in their transcription. Press Enter or click **Search**.                                                                                                                                                               |
| 4 | **Search in:**                | **Project & file name** or **Transcribed text**. A text search shows how many matches each page has.                                                                                                                                                    |
| 5 | **All**                       | Selects every page on the list page.                                                                                                                                                                                                                    |
| 6 | **Grid view** / **List view** | Switches between cards and rows.                                                                                                                                                                                                                        |
| 7 | **Upload**                    | Starts a new project.                                                                                                                                                                                                                                   |
| 8 | **Refresh**                   | Reloads the statuses. The list also refreshes itself every few seconds while recent pages are still transcribing.                                                                                                                                       |
| 9 | Status pill                   | **Not started**, **Queued**, **In progress**, **Completed**, or **Failed**. A **Verified** pill marks checked pages, and **NEW** marks pages added in the last few minutes.                                                                             |

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-366d420f67a3681f9b682494f1a26d8905ab0d64%2Fsettings-image-to-text-14-card.png?alt=media" alt="A completed manuscript page card with its hover icons and verify toggle"><figcaption><p>Hover a card to show its icons; a greyed icon names the reason in its tooltip.</p></figcaption></figure>

Each card shows the page number, file name, project, language, and last change, plus these icons: **Quick preview** (with **Open transcription**), **Download image**, **Download DOCX**, **Download spreadsheet**, **Copy text**, and **Delete image**. The circle in the corner is the verify toggle (**Mark as verified** / **Mark as not verified**); it is blocked until the page has finished.

A **Failed** page, or a completed page with no text, shows **Retry**, which runs the page again with the advanced model and the language and domain you chose (not an edited prompt). A **Queued** page does not open until processing begins. Clicking a **Not started** page reopens the wizard so you can finish the setup.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-a82c2ca2cae734d60c79c6bfb39b2d11a0f3a586%2Fsettings-image-to-text-15-selection.png?alt=media" alt="Selection bar with two pages selected and the transcript download menu open"><figcaption><p>Several pages download together as a ZIP.</p></figcaption></figure>

Ticking pages shows a selection bar: **Download image**, a **Download N transcripts** menu (**N separate DOCX files**, **All N in one DOCX**, or a **Spreadsheet** with one row per page; PDF pages can come as one DOCX per PDF), **Delete**, and **Clear selection**.

{% hint style="warning" %}
**Cannot be undone:** **Delete** on the selection bar, **Delete image** on a card, and **Delete file** or **Delete all files** in the viewer remove the pages and their text for good.
{% endhint %}

## The page viewer

Open a page to see it beside its transcription.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-4b57a2b90067089e5dbcdf734c39cbfacf641aae%2Fsettings-image-to-text-16-viewer.png?alt=media" alt="Viewer header with name, ID, progress chips, Spreadsheet, All images, Files, layouts, Delete file and Mark verified"><figcaption><p>The chips under the name count transcribed, verified, and failed pages.</p></figcaption></figure>

| # | Control                                         | What it does                                                                                                                                                          | Notes                                                                                            |
| - | ----------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------ |
| 1 | **Rename project**                              | Edits the project name in place.                                                                                                                                      | Enter saves, Escape cancels.                                                                     |
| 2 | **Copy project ID**                             | Copies the project's ID, for support requests.                                                                                                                        |                                                                                                  |
| 3 | Status and progress chips                       | The project status (**Completed**, **In Progress**, **Not started**, **Failed**, or **Partially Completed**) and how many pages are transcribed, verified, or failed. | A line below says if pages are still transcribing, or were never sent (with **Continue setup**). |
| 4 | **Spreadsheet**                                 | Downloads every page of the project as one workbook, one row per page.                                                                                                | Greyed out until something is transcribed.                                                       |
| 5 | **All images**                                  | Returns to the list (**Search results** if you came from a search).                                                                                                   |                                                                                                  |
| 6 | **Files (N)**                                   | Opens the files panel.                                                                                                                                                |                                                                                                  |
| 7 | **Split view** / **Image only** / **Text only** | The viewer layout.                                                                                                                                                    |                                                                                                  |
| 8 | **Delete file**                                 | Deletes this page from the project.                                                                                                                                   |                                                                                                  |
| 9 | **Mark verified** / **Mark not verified**       | Records your check.                                                                                                                                                   |                                                                                                  |

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-32a5693a1a1735040a4475653f95b16cf744b81b%2Fsettings-image-to-text-17-panes.png?alt=media" alt="Image pane with zoom, rotate and Download beside the text pane with Formatted, Raw, Edit, Copy, DOCX and XLSX"><figcaption><p>The mouse wheel zooms the image and dragging pans it.</p></figcaption></figure>

| #  | Control         | What it does                                                                          | Notes                                                                         |
| -- | --------------- | ------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
| 1  | **Fit to view** | Fits the page in the pane and resets the rotation.                                    |                                                                               |
| 2  | Zoom menu       | Sets the zoom (50% to 300%).                                                          | **Zoom out** and **Zoom in** sit beside it.                                   |
| 3  | **Rotate**      | Turns the image.                                                                      |                                                                               |
| 4  | **Download**    | Downloads the original image.                                                         |                                                                               |
| 5  | **Formatted**   | Shows tables, headings, and line breaks as on the page.                               | Your choice is kept as you move between pages.                                |
| 6  | **Raw**         | Shows the Markdown source with line numbers.                                          |                                                                               |
| 7  | **Edit**        | Opens the text for correction.                                                        | **Save** (Ctrl+Enter or Cmd+Enter) keeps your changes; **Cancel** drops them. |
| 8  | **Copy**        | Copies the text, with formatting where the target accepts it.                         |                                                                               |
| 9  | **DOCX**        | Downloads this page as a Word file with the page image.                               | For a page from a PDF, choose **This page** or **Whole PDF**.                 |
| 10 | **XLSX**        | Downloads a spreadsheet: a table on the page becomes rows, other lines one cell each. |                                                                               |

**Previous page** and **Next page** (or the arrow keys) move through the project; they float over the image when the project has more than one page.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-c380d1abc93bd42edd95db91866ca1d95718eb97%2Fsettings-image-to-text-18-raw.png?alt=media" alt="Text pane in Raw mode showing the Markdown source with line numbers"><figcaption><p>Raw is useful for checking exactly what will be exported.</p></figcaption></figure>

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-3e011f7ec84816a7bfd365fcf9264347458aa2cc%2Fsettings-image-to-text-19-files-panel.png?alt=media" alt="Files panel listing the project&#x27;s two pages beside the viewer"><figcaption><p>With more than six files, the panel adds a search box and status filters.</p></figcaption></figure>

The files panel lists every page. With more than six, **Search files…** and a filter appear: **All files**, **Completed**, **Not verified**, **Verified**, **Queued or in progress**, **Not started**, and **Failed** (only those with pages). **Delete all files** at the bottom deletes every page of the project.

<figure><img src="https://1178311500-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fi4GBXPMUDxszIpUVOARU%2Fuploads%2Fgit-blob-1c63e835562787455774298a6913597649d3ca78%2Fsettings-image-to-text-20-edit.png?alt=media" alt="Editing transcription with the text open for changes and Cancel and Save buttons"><figcaption><p>Leaving with unsaved changes asks whether to discard them.</p></figcaption></figure>

## Next steps

* [Convert Images & PDFs to Text](/ailyze-docs/prepare-research-data/convert-images-and-pdfs-to-text.md) — the walkthrough these controls belong to.
* [Image to Text Issues](/ailyze-docs/troubleshooting/image-to-text-issues.md) — unreadable output, stuck pages, and charges.
* [Document Analysis Settings](/ailyze-docs/reference/settings/document-analysis.md) — what to do with the DOCX files you export.