📝 Overview
Yoom's AI-powered OCR feature allows you to extract text from images and PDFs.
You can extract from various file formats, such as invoices, business cards, and documents.
📌 Important Notes:
Each extraction action has different capabilities and task consumption.
The setup process is the same across all actions.
For more advanced text extraction, we recommend using v2.
💡 Plans Available for Use
AI features (including OCR and summarization) are available in Business plans and above.
Free plan users can access a two-week free trial (once per plan).
Signing up for a Business or Enterprise plan trial grants access to AI features for a limited time.
Example Configuration 1: Read a Invoice (5 tasks)
1. Click the "+" button on the Flowbot edit screen.
2. In the Operation Type selection, click "Read text from images/PDFs."
3. Enter an optional title, set the "Read a Qualified Invoice (5 tasks)" action, and click Next.
4. Attach the file, set any additional items you want to extract if needed, then run a test. If the text is successfully read, save your changes.
*The methods for attaching files are as follows:
Use the acquired values: This method uses the outputs obtained within this Flowbot. (For more information about outputs, please refer to here.)
Upload an image: This method uploads the data directly on this screen.
*For files with a fixed format such as invoices or business cards like this example, default retrieval items are set.
If there is any other text you want to extract, please enter the desired items in the additional extraction items field as in the example below.
Example: Bank name, invoice title, etc.
Example Configuration 2: Read Text Information (up to 6,500 characters) from PDF or Image Files (15 tasks)
*The operation steps up to Step 2 in Example Configuration 1 are the same.
3. Enter an optional title, set the "Read text information from PDF or image files (15 tasks)" action, and click Next.
4. Attach the file, configure any additional items you want to extract if needed, then run a test. If the text is successfully read, save your changes.
*The methods for attaching files are as follows:
Use the acquired values: This method uses the outputs obtained within this Flowbot. (For more information about outputs, please refer to here.)
Upload an image: This method uploads the data directly on this screen.
*In the "Read text information from PDF or image files (15 tasks)" action,
there are no default text extraction settings,
so you need to enter the items you want to extract.
Example: The text of Article 2, etc.
Sample Flowbot
Notes
The number of tests allowed varies by action. If you exceed the limit,
please wait until the next available test time.
There may be instances where reading the entire file's text or text across multiple lines may not be read properly due to specifications.
Please be aware of this.
The maximum file size that can be read is 7MB.
Please limit your files to 7MB or less.
Due to AI specifications, there is a certain likelihood of reading failure.
In such cases, please run it again.
Related Help Pages
That concludes the configuration method for "Extracting text from images/PDFs."
# PDF, file, image, reading, OCR, reading PDF or images, transcription, extraction, text