Skip to main content


The Postprocess endpoint performs a variety of operations that are useful for working with the outputs of an NLP (whether human or AI) system. These include:

  • Fixing or restoring punctuation
  • Fixing capitalization
  • Fixing or restoring whitespace
  • Splitting sentences into shorter sentences

Example use cases include repairing transcriptions or captions.

Looking for something not provided here? Contact us.

Postprocess POST

Request Parameters

HTTP method: POST

The Postprocess API POST endpoint takes JSON parameters documented below:

key: String
32-character API key.

text: String
Text to postprocess

session_id: String
Unique name or UUID of document or portion of text that is being checked.

operations: String List
Operations to apply. The currently accepted operations are:

  • capitalize
  • punctuate
  • fixspaces
  • split (Beta: please contact us if you encounter any issues)

auto_apply: Optional Boolean
Default is false. If true, result with have extra field applied_text containing text with edits applied.

Example JSON body:

"key": "<api-key>",
"text": "Hey how are you doing",
"session_id": "test_session",
"operations": ["capitalize", "punctuate"]

Response Parameters

The Postprocess endpoint returns JSON following the same structure as the edits endpoint.

edits: Array
Array of Edit objects for the text sent in the request.

Example output:

"edits": [
"end": 3,
"error_type": "M:PUNCT",
"general_error_type": "Punctuation",
"id": "b8d79060-d244-59ef-8dc8-9fdd02216454",
"replacement": "Hey,",
"sentence": "Hey how are you doing",
"sentence_start": 0,
"start": 0
"end": 21,
"error_type": "M:PUNCT",
"general_error_type": "Punctuation",
"id": "5b5f7598-91ef-55db-9359-0c88e11e5259",
"replacement": "doing?",
"sentence": "Hey how are you doing",
"sentence_start": 0,
"start": 16