curl --request POST \
  --url https://api.getlimina.ai/community/v4/ner/files/base64 \
  --header 'Content-Type: application/json' \
  --header 'x-api-key: <api-key>' \
  --data '
{
  "file": {
    "data": "JVBERi0xLjQKJdPr6eEKMSAwIG9iago8PC9UaXRsZSAoc2FtcGxlKQovUHJvZHVj...",
    "content_type": "application/pdf or image/jpeg"
  },
  "entity_detection": {
    "return_entity": true
  },
  "pdf_options": {
    "density": 150,
    "max_resolution": 2000
  },
  "audio_options": {
    "bleep_start_padding": 0,
    "bleep_end_padding": 0
  },
  "ocr_options": {
    "ocr_system": "azure_computer_vision"
  }
}
'

{
  "entities": [
    {
      "text": "<string>",
      "location": {
        "stt_idx": 123,
        "end_idx": 123,
        "page": 0
      },
      "label": "<string>",
      "likelihood": 123,
      "gibberish_score": 0.5
    }
  ],
  "entities_present": true,
  "languages_detected": {},
  "characters_processed": 123,
  "extracted_text": "<string>",
  "objects": [
    {
      "type": "FACE",
      "location": {
        "page": 123,
        "x0": 0.5,
        "x1": 0.5,
        "y0": 0.5,
        "y1": 0.5
      }
    }
  ],
  "objects_present": false,
  "audio_duration": 123,
  "page_count": 123
}

{
  "detail": "<string>"
}

{
  "detail": "<string>"
}

{
  "detail": "<string>"
}

File

Ner Files Base64

Detect entities such as PII, PHI or PCI in a base64-encoded file using Private AI’s entity detection engine. This is only doing the detection of entities, no redacted file is created.

This route is similar to /ner/files/uri, but passes the file in the POST request itself. This route allows for simple setup and testing, as no folders or volumes need to be mounted to the container.

Set return_extracted_text to True to also receive the full text extracted from the file (the OCR output for images and scanned PDFs) in the extracted_text field. Note that this extracted text is not de-identified.

This route supports the following content types: application/pdf, application/vnd.openxmlformats-officedocument.wordprocessingml.document, image/bmp, image/gif, image/jpeg, image/jpg, image/png, image/tif, image/tiff, image/x-ms-bmp, text/csv, text/plain

POST

ner

files

base64

curl --request POST \
  --url https://api.getlimina.ai/community/v4/ner/files/base64 \
  --header 'Content-Type: application/json' \
  --header 'x-api-key: <api-key>' \
  --data '
{
  "file": {
    "data": "JVBERi0xLjQKJdPr6eEKMSAwIG9iago8PC9UaXRsZSAoc2FtcGxlKQovUHJvZHVj...",
    "content_type": "application/pdf or image/jpeg"
  },
  "entity_detection": {
    "return_entity": true
  },
  "pdf_options": {
    "density": 150,
    "max_resolution": 2000
  },
  "audio_options": {
    "bleep_start_padding": 0,
    "bleep_end_padding": 0
  },
  "ocr_options": {
    "ocr_system": "azure_computer_vision"
  }
}
'

{
  "entities": [
    {
      "text": "<string>",
      "location": {
        "stt_idx": 123,
        "end_idx": 123,
        "page": 0
      },
      "label": "<string>",
      "likelihood": 123,
      "gibberish_score": 0.5
    }
  ],
  "entities_present": true,
  "languages_detected": {},
  "characters_processed": 123,
  "extracted_text": "<string>",
  "objects": [
    {
      "type": "FACE",
      "location": {
        "page": 123,
        "x0": 0.5,
        "x1": 0.5,
        "y0": 0.5,
        "y1": 0.5
      }
    }
  ],
  "objects_present": false,
  "audio_duration": 123,
  "page_count": 123
}

{
  "detail": "<string>"
}

{
  "detail": "<string>"
}

{
  "detail": "<string>"
}

Authorizations

x-api-key

string

header

required

Body

application/json

file

File · object

required

Base64 encoded file content.

Show child attributes

entity_detection

PIIDetectionParams · object

This section contains a set of parameters to control the PII detection process. All fields have sensible default that can be changed for specific needs.

Show child attributes

object_entity_detection

ObjectEntityDetection · object

This section contains a set of parameters to control the object entity detection process. It allows the user to select the object entity types to detect (e.g., to detect FACE but not LICENSE_PLATE).

Show child attributes

pdf_options

PDFOptions · object

Options to process PDF files, such as the rendering quality when each page is turned into an image.

Show child attributes

office_options

OfficeOptions · object

Options to process Office files, such as table and chart behaviour.

Show child attributes

image_options

ImageOptions · object

Options to process image files, such as the masking mode.

Show child attributes

audio_options

AudioOptions · object

Options to process audio files, such as the padding to add while redacting audio segments.

Show child attributes

project_id

string

default:main

Used to categorize requests for reporting purposes. Limited to alphanumeric characters or the following special characters :_-

Maximum string length: 60

Pattern: ^[a-zA-Z0-9\-_\:]*$

ocr_options

AWSTextractOCROptions · object

Options to provide Optical Character Recognition (OCR) details, such as choice of OCR system.

AWSTextractOCROptions
AzureComputerVisionOCROptions
AzureDocIntelligenceOCROptions
PaddleOCROptions
HybridOCROptions

Show child attributes

enable_gibberish_detection

boolean

default:false

When set to True, the gibberish detector is enabled, returning a gibberish_score for each entity in the response.

return_extracted_text

boolean

default:false

Controls whether the response contains the extracted_text field, which holds the full text extracted from the input file (via OCR for images and scanned PDFs) before any redaction is applied. Note that the extracted text is not de-identified, so any PII it contains is returned as is, independently of the entity_detection.return_entity setting. Turning this ON can significantly increase the size of the response, so it defaults to False.

Response

Successful Response

entities

(NerFileEntityItem · object | NerOfficeFileEntityItem · object)[]

required

A list of all entities found in the provided file.

Empty

NerFileEntityItem
NerOfficeFileEntityItem

Show child attributes

entities_present

boolean

required

Returns True if the list of detected entities is not empty.

languages_detected

Languages Detected · object

required

A dictionary containing ISO 639-1 language labels and the likelihood of their detection in the request payload.

Show child attributes

characters_processed

integer

required

The number of characters extracted from the file.

extracted_text

string | null

The full text that was extracted from the input file, before any redaction is applied. It corresponds to the OCR output for image files and scanned PDFs, and to the file contents for text-based formats such as .txt and .csv. Note that this text is not de-identified, so any PII it contains is returned in the clear.

This field is only present if the return_extracted_text option is set to True in the request.

objects

FileObjectEntityItem · object[]

A list of all object entities found in the provided file using object detection.

Show child attributes

objects_present

boolean

default:false

Returns True if the list of detected objects is not empty.

audio_duration

number | null

The length of the audio file in seconds.

page_count

integer | null

The number of pages in the file.