Uploading Resources
Upload documents, spreadsheets, presentations, and more to train your companion.
Resources are the files you upload to give your companion deep expertise. The companion processes these files, learns from them, and uses that knowledge to answer user questions accurately.
Upload Flow
Open the Uploader
In the Data tab, find the Resources section at the bottom. Click the Upload button to open the uploader.
Add Your Files
You can drag and drop files into the upload area, or click Browse to select files from your device. You can upload up to 50 files at once.
The uploader accepts all supported document formats, videos, and images. Each file shows a format-specific icon so you can easily identify what you're uploading.
Configure OCR (Optional)
If you are uploading scanned documents, images with text, or documents with embedded images of text, you can enable OCR (Optical Character Recognition) to extract text from images within the document.
- Toggle OCR on or off for each file
- Select the language of the text in the document (see the supported languages table below)
Upload and Process
Click Upload to start. Your files will be uploaded and then queued for processing. You can track the progress in the progress tracker that appears at the top of the page.
Supported File Types
Documents
| Type | Extensions |
|---|---|
| Microsoft Word | .doc, .docx, .docm |
| Microsoft Excel | .xls, .xlsx, .xlsm |
| Microsoft PowerPoint | .ppt, .pptx, .pptm |
| OpenDocument Text | .odt |
| OpenDocument Spreadsheet | .ods |
| OpenDocument Presentation | .odp |
| CSV / TSV | .csv, .tsv |
| Markdown | .md |
| Plain Text | .txt |
All office formats (Word, Excel, PowerPoint, OpenDocument) are automatically converted and parsed. CSV, TSV, Markdown, and plain text files are parsed natively.
Videos
| Type | Extensions |
|---|---|
| MP4 | .mp4 |
| WebM | .webm |
| Ogg | .ogg |
Videos are automatically transcribed -- the companion extracts the spoken content and learns from it.
Images
| Type | Extensions |
|---|---|
| JPEG | .jpg, .jpeg |
| PNG | .png |
| WebP | .webp |
| GIF | .gif |
Images are automatically analyzed when uploaded. Aura's AI examines each image using vision analysis to generate a descriptive caption, and optionally runs OCR to extract any text in the image. The companion uses this understanding to answer questions about the image content. Images can also be displayed in conversations (via the Rich Media skill) and in the Featured Photos gallery.
OCR Languages
If you enable OCR on a document, you can choose from 15 supported languages:
| Language | Code |
|---|---|
| English | en |
| Chinese (Simplified) | zh |
| Chinese (Traditional) | zh-tw |
| Japanese | ja |
| Korean | ko |
| French | fr |
| German | de |
| Spanish | es |
| Portuguese | pt |
| Italian | it |
| Russian | ru |
| Arabic | ar |
| Hindi | hi |
| Dutch | nl |
| Greek | el |
The default language is English. Choose the language that matches the text in your document for best results.
Processing Status
After uploading, each resource goes through a series of stages:
| Status | Meaning |
|---|---|
| Pending | The file has been uploaded and is waiting to be queued |
| Queued | The file is in the processing queue |
| Processing | The file is actively being analyzed and learned from |
| Ready | Processing is complete -- the companion can now use this resource |
| Failed | Something went wrong during processing |
Processing Stages
During processing, each document goes through these stages (visible in the progress tracker):
- Preparing — downloading and setting up the file for processing
- Converting — converting office formats for parsing (Word, PowerPoint, Excel, OpenDocument formats only)
- Scanning — extracting text content with optional OCR
- Learning — embedding the content and indexing it for AI retrieval
- Finalising — updating the resource status and completing the job
Image Processing
When an image is processed, two things happen:
- AI Vision Analysis — The image is analyzed by an AI vision model that generates a detailed caption describing the image content, objects, text, and context.
- OCR (if enabled) — Optical Character Recognition extracts any readable text from the image.
The generated caption and extracted text become part of the companion's knowledge, allowing it to answer questions about the image intelligently.
Browsing Your Resources
Resources are organized into three tabs:
- Documents -- PDFs, Word, Excel, PowerPoint, OpenDocument, CSV, Markdown, and text files
- Videos -- MP4, WebM, and Ogg files
- Images -- JPEG, PNG, WebP, and GIF files — each with AI-generated captions viewable in the image viewer
Each tab shows a count of how many resources you have (e.g., "Documents (12)").
Reprocessing a Resource
If a resource failed to process or you want to reprocess it (for example, with different OCR settings), you can click the reprocess option on any resource. This clears the previous processing results and re-queues the file. For images, reprocessing regenerates the AI caption and re-runs OCR if enabled. You can change OCR settings (enable/disable OCR, change language) when reprocessing.
Image Viewer
Click any processed image to open the Image Viewer. The viewer shows:
- The full-size image
- The AI-generated caption describing the image content
- Controls to edit, regenerate, or delete the caption
- OCR status and language settings
- Processing status indicator
You can regenerate captions with different OCR settings if the initial result was not satisfactory.
Processing can take some time, especially for large documents and videos. The speed depends on the file size and complexity. You can continue working on other parts of your companion while files process in the background.
Knowledge vs. Resources -- Knowledge sections are quick facts you type manually and are great for FAQs and structured data. Resources are files the companion reads and learns from in depth. For the best results, use both: upload detailed documents as resources and add key highlights as knowledge items.
Last updated on