Attachment Transcript

Overview

Ever had a 50-page supplier catalog and wished you could search for that one specific technical spec? {productName} can now “read” your attachments for you. This includes images, scanned PDFs, and even voice recordings. Once the system finishes reading a file, you can find anything inside it using the regular search bar. This helps you find products faster and gives the AI more information to work with when it helps you write your product descriptions.

[navigate:/products|Products|shopping-cart]

How it works

1. Go to **Settings** > **AI Settings** and choose a mode in **Attachment Transcript to Text**: **AI transcription** reads everything, including images, audio and scanned PDFs, by sending the files to the AI provider; **Local extraction only** keeps all your files on the server and extracts just the text already present in them (text files, Word, Excel, PowerPoint, PDFs with a text layer); **Off** disables reading attachments.
2. Select the products you want to work with in the **Products** grid.
3. Open the **Attachments** panel on the right side of the editor.
4. Upload your supplier PDF, product manual, or a photo of a spec sheet.
5. Wait a few moments while the system reads the content of your files in the background.
6. Use the global search bar at the top or the grid filters to search for any word or phrase found inside those documents.

What can it read?

card: Scanned Documents
icon: file-text
The system reads text from PDFs and Word documents, even if they were scanned from paper.

card: Images & Photos
icon: camera
Upload a photo of a product label or a technical drawing to make the text on it searchable.

card: Audio & Voice
icon: mic
If you record a short voice note about a product, the system will turn it into a searchable description.

Tips

**Tip:** Use this to handle supplier specification sheets. Once the text is extracted, you can use **Smart Bulk Updates** to quickly fill in missing attributes like "Material" or "Dimensions" based on the data the system found in the file.

Troubleshooting

**Nothing found inside an image, a scan or a voice note?** With **Local extraction only** the system never sends files to the AI provider, so it can only read text that already exists in the file. Switch to **AI transcription** if you want photos, scanned documents and audio to be read too.

[navigate:/help/en/user-flows/ai-mass-update|Smart Bulk Updates with AI|zap]
[navigate:/help/en/user-flows/image-similarity-search|Image Similarity Search|image]
[navigate:/help/en/user-flows/bulk-image-management|Bulk Image Management|image]
[navigate:/help/en/ai-credit|What Is AI Credit?|sparkles]