AI automation
for your software

AI document classification

Why natif.ai

Position data

In addition to classic header data, our AI can also extract position data.

Accuracy

Our AI models have been trained with hundreds of thousands of documents and extract all important information with high precision.

Speed

Thanks to high-performance GPUs, our AI processes documents in real time.

Swarm intelligence

Our AI benefits from feedback from all our customers, so problems don’t have to be solved by each customer anew.

No templates

We do not use outdated template-based algorithms, but work with state-of-the-art Deep Learning approaches.

Robust reading

Our Deep OCR reads even low-quality documents and can be specially adapted to difficult use cases.

Data protection

We are 100 % GDPR- and Schrems II-compliant. We process the data on our servers and dispense with service providers.

AI Boost

Our AI models directly enable your software to automate documents.

Optical Character Recognition (OCR)

Organisation

NatifSports

Entry date

01.01.2023

Date of birth

13.06.1990

Mobile number

+49 151 687899

Signature

M. Mustermann

{
  "processing_id": "bfb819fb-b193-42fa-ad19-8ad33aa8831b",
  "available_results": [
    "pdf",
    "ocr",
    "page-images",
    "hocr",
    "language",
    "thumbnail"
  ],
  "pdf": "/processing/results/bfb819fb-b193-42fa-ad19-8ad33aa8831b/pdf",
  "ocr": {
    "pages": [
      {
        "width": 2380,
        "height": 3368,
        "bboxes": [
          {
            "id": 1,
            "x1": 82,
            "y1": 333,
            "x2": 102,
            "y2": 351,
            "text": "a",
            "text_entropy": 0.58203125
          },
          {
            "id": 2,
            "x1": 132,
            "y1": 1508,
            "x2": 551,
            "y2": 1558,
            "text": "Staatsangehörigkeit:",
            "text_entropy": 0.141357421875
          },
          {
            "id": 3,
            "x1": 152,
            "y1": 108,
            "x2": 5551,
            "y2": 1558,
            "text": "Aktives Mitglied in der Sparte",
            "text_entropy": 0.161421875
          },
....
Show more +

Optical Character Recognition (OCR)

Optical Character Recognition (OCR) is the cornerstone of enterprise document process automation. Our OCR uses innovative deep learning and efficiently delivers high-quality results. It recognizes even illegible scans of crumpled documents as well as handwritten documents with high precision.

Buchhaltungs- Automatisierung

Receiver address

Erfurterstrasse 13

Sender

JOKO Service

Sender address

Lebacher Sand 11

Invoice No.

M1675

Invoice date

13.03.2019

Article No.

B-3025-078

Total article “B-3025-078”

120,00

Total invoice

209,03

Tax No.

91/200/53688

{
  "schema_version": 1,
  "document_type": "invoice",
  "customer": {
    "name": {
      "validation_problem": false,
      "note": "",
      "confidence": 0.99985,
      "bbox_refs": [
        {
          "page_num": 1,
          "bbox_id": 34
        },
        {
          "page_num": 1,
          "bbox_id": 35
        }
      ],
      "value": "Klaus Gotti"
    },
    "address": {
      "validation_problem": false,
      "note": "",
      "confidence": 0.999925,
      "bbox_refs": [
        {
          "page_num": 1,
          "bbox_id": 36
        },
        {
          "page_num": 1,
          "bbox_id": 37
        },
        {
          "page_num": 1,
          "bbox_id": 30
        },
        {
          "page_num": 1,
          "bbox_id": 31
        }
      ],
      "value": "Erforterstrasse 13 66111 Saarbrücken"
....
Show more +

Accounting automation

Accounting documents such as invoices, delivery notes and receipts are among the most common business documents. Any business software should be able to automate such important and everyday documents.
In addition to information such as details of the sender and recipient, our software can also read out item data, tax rates and discounts.

Ausweis-Dokumente

Ausweisnr.

LZ6311T47

Nachname

Mustermann

Geburtsname

Gabler

Vorname

Erika

Geburtstag

2019-03-13

Staatsangehörigkeit

Deutsch

Geburtsort

Berlin

Gültig bis

31.10.2031

Augenfarbe

Grün

Größe

160 cm

{
  "processing_id": "e454fa71-fbe2-4a23-8a8a-7db196dd114c",
  "available_results": [
    "ocr",
    "page-images",
    "hocr",
    "language",
    "thumbnail"
  ],
  "ocr": {
    "pages": [
      {
        "width": 2179,
        "height": 1379,
        "bboxes": [
          {
            "id": 1,
            "x1": 192,
            "y1": 102,
            "x2": 285,
            "y2": 156,
            "text": "DE",
            "text_entropy": 0.00545501708984375
          },
          {
            "id": 2,
            "x1": 445,
            "y1": 192,
            "x2": 571,
            "y2": 212,
            "text": "IDENTITY",
            "text_entropy": 0.07196044921875
....
Show more +

Identity documents

KYC (know your customer) identification is an essential onboarding process. Thanks to our AI technology, IDs can be found in documents, trimmed and read in real time. All relevant data points can be transmitted via our Rest API. Despite common anti-counterfeiting features, our AI can read the relevant content with ease.

Dokumenten-Klassifizierung

Document type

Delivery note

Receiver name

Alfred Schneider

Receiver address

Muttweg 3

Date

01.05.2022

Delivery No.

3931

Article No. (1)

2912-282

Description (1)

Hook-M/3

Vendor

Triskona GmbH

Vendor VAT ID

DE 4567883213

{
  "processing_id": "e0e053a8-9efe-4a40-a5b3-758b69da39bb",
  "available_results": [
    "extractions",
    "classifications",
    "pdf",
    "ocr",
    "page-images",
    "hocr",
    "language",
    "thumbnail"
  ],
  "extractions": {
    "schema_version": 2,
    "document_type": "delivery note",
    "customer": {
      "name": {
        "validation_problem": false,
        "note": "",
        "confidence": 0.99985,
        "bbox_refs": [
          {
            "page_num": 1,
            "bbox_id": 3
          },
          {
            "page_num": 1,
            "bbox_id": 21
          }
        ],
        "value": "Alfred Schneider"
      },
      "address": {
        "validation_problem": false,
        "note": "",
        "confidence": 0.9998,
        "bbox_refs": [
          {
            "page_num": 1,
            "bbox_id": 10
          },
          {
            "page_num": 1,
            "bbox_id": 11
          },
          {
            "page_num": 1,
            "bbox_id": 1
          },
          {
            "page_num": 1,
            "bbox_id": 2
          }
        ],
        "value": "Mutweg 3 66117 Saarbrücken"
      },
...
Show more +

Document type

Rejection legal expenses insurance

Receiver name

Rechtsanwaltskanzlei Müller

Receiver address

Petersstraße 1

Receiver ZIP

49808

Receiver city

Lingen

Date

24.03.2021

Sender name

Saar Direct Versicherung

Sender address

Postfach 21 09 07

Subject

 Deckungsanfrage

{
  "processing_id": "220a05b5-a313-46dc-9b53-7db32a15eced",
  "available_results": [
    "extractions",
    "pdf",
    "ocr",
    "page-images",
    "hocr",
    "language",
    "thumbnail"
  ],
  "extractions": {
    "schema_version": 1,
    "document_type": "letter",
    "receiver": {
      "name": {
        "validation_problem": false,
        "note": "",
        "confidence": 0.9998,
        "bbox_refs": [
          {
            "page_num": 1,
            "bbox_id": 92
          },
          {
            "page_num": 1,
            "bbox_id": 93
          }
        ],
        "value": "Rechtsanwaltskanzlei Müller"
      },
      "address": {
        "address_line_1": {
          "validation_problem": false,
          "note": "",
          "confidence": 0.99985,
          "bbox_refs": [
            {
              "page_num": 1,
              "bbox_id": 94
            },
            {
              "page_num": 1,
              "bbox_id": 95
            }
          ],
          "value": "Petersstraße 1"
        },
....
Show more +

Document type

Invoice

Receiver address

Erfurterstrasse 13

Sender name

JOKO Service

Sender address

Lebacher Sand 11

Invoice No.

M1675

Invoice date

13.03.2019

Article No. (1)

B-3025-078

Price total (1)

120,00

Tax No.

91/200/53688

{
  "schema_version": 1,
  "document_type": "invoice",
  "customer": {
    "name": {
      "validation_problem": false,
      "note": "",
      "confidence": 0.99985,
      "bbox_refs": [
        {
          "page_num": 1,
          "bbox_id": 34
        },
        {
          "page_num": 1,
          "bbox_id": 35
        }
      ],
      "value": "Klaus Gotti"
    },
    "address": {
      "validation_problem": false,
      "note": "",
      "confidence": 0.999925,
      "bbox_refs": [
        {
          "page_num": 1,
          "bbox_id": 36
        },
        {
          "page_num": 1,
          "bbox_id": 37
        },
        {
          "page_num": 1,
          "bbox_id": 30
        },
        {
          "page_num": 1,
          "bbox_id": 31
        }
      ],
      "value": "Erforterstrasse 13 66111 Saarbrücken"
    },
....
Show more +

Document classification

Classifying documents has been a major challenge for a wide variety of software vendors. Our classification technology can classify documents into different classes as well as different business transactions.
It is ready to use without any setup. On our platform, there is an additional possibility to train your own document classification for specific problems.

We care about our customers

Will Ellis– CTO
“Integrating our product with Natif’s powerful AI and OCR technology was a smooth process, which has already borne fruit by reducing our team’s heavy manual workload. We are excited to continue the partnership and automate the data extraction from more document types.”
show more
Dr. Dirk Goldner– board member
“With natif.ai, we have found a reliable partner. The technology provides us with excellent support in recognizing the essential document information that is important for further processing in our solutions – in both the German and international context. For us, this makes the difference in significantly increasing the current level of automation.”
show more
Michael Schlumberger– managing director
“The integration of natif.ai was very easy, the AI runs in the background and quickly transmits the extraction data to our arkivado TURM (transaction server) and prepares the data for the ecoDMS document management system. Thus, our customers receive a fully automated classification of their incoming documents via email forwarding (autom. or manual). Thanks to the excellent data extraction, we can also implement extensive automated workflows. Since natif.ai processes the data exclusively in Germany, the entire process is GDPR-compliant – which is currently not common in this area.”
show more
Rainer Schramm– managing director
“With natif.ai, we automate the capture and digitization of our customers’ invoices. Thanks to OpenAPI, the accounting workflow was quick and unproblematic to connect to the ERP system weclapp. Through automated document classification and extraction of key data, we reduce errors and manual workload.”
show more
Tobias Hoffmann– director business solutions
“Thanks to natif.ai, we enable our customers to automate their invoice receipt including the position data – and without high implementation costs. In addition, it is now possible for us to classify incoming mail customer-specifically without much effort. The integration of natif.ai was quick and straightforward, so that we were able to automate our customers’ document processing within a short time and at low cost. Since natif.ai’s document extraction works directly and without complex system configuration, it offers an excellent option for digitizing medium-sized companies.”
show more