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

# Batch import

> Upload multiple files at once. Bower processes each one with the right model and turns them into searchable, editable notes.

Upload multiple files at once. Bower processes each one with the right model and turns them into searchable, editable notes.

## What you can import

Bower handles mixed batches — you don't need to separate files by type before uploading.

| File type                           | What Bower does                                                                                                                                                                                                |
| ----------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Images (.jpg, .png, .heic)          | Generates an AI description of the image and extracts any text it finds. Both the description and extracted text become a searchable note — so you can find images by what's in them, not just their filename. |
| Voice recordings (.m4a, .mp3, .wav) | Transcription with scientific vocabulary recognition                                                                                                                                                           |
| PDFs                                | Content extraction; auto-generates flow charts from embedded images                                                                                                                                            |
| Word documents (.docx)              | Converted to editable notes; formatting including tables and coloured text is preserved                                                                                                                        |

## How to import

<Steps>
  <Step title="Open the import dialog">
    Click the **New note** dropdown in any collection and select **Import**, or use the **+ New note** button on the dashboard.
  </Step>

  <Step title="Select your files">
    Select all files — mixed types are fine. You can import up to **20 files** at a time, with a maximum of **50 MB per file**.
  </Step>

  <Step title="Bower processes them">
    Bower queues each file and processes them in parallel. Each file becomes a separate note with the source file attached.
  </Step>
</Steps>

## Limits

| Limit            | Value          |
| ---------------- | -------------- |
| Files per import | 20             |
| Max file size    | 50 MB per file |

## After import

* Each note is immediately searchable.
* Images are searchable by their AI-generated description — you can find a photo of a gel by searching for "gel electrophoresis" even if the filename is `IMG_4382.jpg`.
* For photos taken on a phone, Bower reads embedded metadata from the file: capture time, GPS location, and device model are shown in the attachment panel when present. Tap the location to open it in Maps.
* Extracted tables are editable inline.
* The original file is always attached to the note — nothing is discarded.
* Bird can summarise, compare, or cross-reference the imported notes as soon as processing completes.

## A note on quality

Text extraction accuracy on printed text is approximately 98%. Handwritten content and instrument displays are lower — Bird will flag low-confidence extractions for your review rather than presenting them silently as verified data. You always have the final say.