Pull fields and tables from documents during a call with Nanonets OCR.

Nanonets OCR turns invoices, forms and other documents into structured data. Connect your Nanonets account to AgentCalls and your agent can look up documents it has already processed, check which models are ready, and extract fields from a document link while a caller is on the line. Callers can ask about an invoice total or a form status, and the agent answers from the data, not from memory. Once connected, the agent uses these actions in the middle of a call, so the answer comes back in the same conversation.

What your agent can do with Nanonets OCR

  • Read extracted fields and tables for a processed document or page
  • Extract fields from a public document link during the call
  • Check which OCR models are ready and what they extract
  • List recent OCR predictions for a model within a date range
  • Confirm a document was processed before the agent answers a question about it

Nanonets OCR exposes 6 actions. You choose which ones the agent may use and whether each runs during or after the call.

Example calls

Invoice status question from a supplier

  1. Caller

    Hi, I'm calling about invoice 4471 from Brightline Supplies. Has it been processed yet?

  2. Agent

    Let me check that for you. Yes, it was processed yesterday afternoon. The total is $3,284.50 and it's due on the 30th.

  3. Caller

    Perfect, thanks. Can you tell me the PO number on it?

  4. Agent

    The PO number on that invoice is PO-20931. Do you want me to send the invoice copy to your email?

Document link sent during a call

  1. Caller

    I just emailed you the signed lease link. Can you pull the start date and monthly rent?

  2. Agent

    Got it. Give me a moment to read the document. The lease starts on March 1 and the monthly rent is £1,450.

  3. Caller

    Great. Is the deposit listed too?

  4. Agent

    The deposit is £2,900. I've saved those fields to your file so your team has them too.

Outbound follow-up on missing form fields

  1. Agent

    Hello, this is calling from Harbor Dental. We received your intake form but the insurance member ID field was blank. Is this a good time to confirm it?

  2. Caller

    Sure, it's HX-7720-binds. Sorry, I missed that.

  3. Agent

    Thank you, I've added HX-7720-binds to your record. Your appointment on Thursday at 10:30 is still confirmed.

How to connect Nanonets OCR

  1. 1

    Create your agent

    Sign up to AgentCalls, pick an English or Arabic voice and write what the agent should know and do in plain language.

  2. 2

    Connect Nanonets OCR

    From Integrations in the dashboard, choose Nanonets OCR and enter the connection details from your Nanonets OCR account settings.

  3. 3

    Choose the actions

    In the agent editor, enable the Nanonets OCR actions the agent may use and set each to run during the call or afterwards. Then test it from your browser.

Questions, answered.

Does it work during the call or after it?

Both. Your agent can look up documents and extract fields while a caller is on the line, and it can also check results after the call when you run follow-up tasks.

What do I need to connect Nanonets OCR?

You need a Nanonets account with access to the OCR models you want to use. Sign in from the AgentCalls dashboard, open Tools, choose Nanonets OCR and enter your credentials from the Nanonets settings page.

Is Nanonets OCR included in my plan?

Tool integrations are included from the Growth plan. On a lower plan, upgrade from your billing page to turn them on.

Page drafted from the tool's public description and reviewed automatically for accuracy of claims

Put your agent to work with Nanonets OCR.

Connect in minutes, test from your browser, go live on a number in the US, UK or Gulf. Free to start, no card needed.