Take the Text in an Image (ChatGPT)

In short

• How to start: In the action search bar, type ChatGPT, and select Take the text in an image.
• Keep in mind: You also can use the Attachment Vision (OCR) feature, instead of this automation.

When to use

• You receive images or scanned files containing text that needs to be stored in a structured field
• You want to digitize handwritten notes, receipts, forms, or labels automatically
• You need to extract text from attachments without manual transcription

---

How to configure

Step 1

Create a new automation
This page covers only the action. For instructions on how to access automations and create a new automation, see the Automations page.

Step 2

Choose the action: In the action search bar, type ChatGPT, and select Take the text in an image.

Step 3

Connect with OpenAI: This action requires an OpenAI connection. Click Connect with OpenAI and follow the authentication steps to link your OpenAI account.

Step 4 (optional)

Choose tab: Select the table where the target record is located.

Step 5

ID of your record: Set the ID of the record you want to update with the extracted text. You can pass this dynamically using a trigger value.

Step 6 (optional)

Choose the field where the text will be saved: Select the text field in which the extracted content will be stored.

Step 7

Image: Provide the content of an image or file field from Jestor. If a file field is used, the first file in the field will always be processed.

Step 8

Save: Click Save to activate the automation.

---

Keep in Mind

• You also can use the Attachment Vision (OCR) feature, instead of this automation
• This action does not process multiple images in a single run: if the field contains more than one file, only the first file is used.
• It does not return structured data: the output is raw extracted text, not parsed or formatted values.
• It does not self-correct OCR errors: if the image has low quality or poor contrast, the extracted text may be inaccurate.
• It does not save the result automatically unless a target field and record ID are properly configured.
• Each execution consumes OpenAI API credits: high-volume automations may incur significant API costs depending on your OpenAI plan.

---

FAQ

1. Does this action work with handwritten text?
Yes. ChatGPT's vision model handles handwritten text well, but accuracy depends on image quality. Clear, well-lit images produce better results.

2. Is there a cost to use this action?
Yes. This action uses the OpenAI API, which is billed separately based on your OpenAI account plan. Jestor does not cover OpenAI API costs.

3. Can I use the extracted text in the same automation to trigger other actions?
Yes. Once the text is saved to a field, subsequent automation steps or separate automations can reference that field value normally.