🖼️ How to Transcribe Images?
1. What Is This New Feature?
Stevo can now automatically read and summarize images and PDFs that users send through the system.
| Type | What the AI Does |
|---|---|
| Images | Describes what appears in the photo (objects, people, context, etc.) |
| PDFs | Generates a summary of the content (e.g. "payment receipt, amount X, date Y, bank Z…") |
This helps:
- ✅ Operations that receive many documents/receipts/proofs
- ✅ Bots and agents understand what the customer sent without opening the file
- ✅ AI flows (assistants, bots) “see” the image/PDF through the summary
2. Requirements
Before enabling:
- ✅ Have a valid ChatGPT API Key (OpenAI)
- ✅ Save this key inside Stevo in the AI / ChatGPT settings
- ✅ (Optional / future) Have keys from other providers like Groq once available
Important
At the moment, the feature works only with ChatGPT. Groq and others will be added later.
3. How to Enable Image and PDF Summaries in Stevo
3.1. Accessing Settings
- ✅ In the Stevo panel, click Settings
- ✅ Enter General / CRM Settings (or your equivalent section)
3.2. Enabling the Features
Inside Settings, you will see two new options:
- Summarize images
- Summarize PDFs
Steps:
- ✅ Locate the switches: “Summarize Images” and “Summarize PDFs”
- ✅ Enable them (you can activate only images, only PDFs, or both)
- ✅ Choose which AI will generate the summary (currently: ChatGPT)
- ✅ Confirm that your ChatGPT API key is filled and saved
4. How It Works in Practice
After activation:
4.1. Sending an Image
- The user sends an image in the conversation
- The system sends that image to the AI
- The AI returns a text description, which appears on screen
dica