Speech Understanding

PII Redaction

Task Description

The PII Redaction endpoint identifies and redacts personally identifiable information (PII) from text transcripts. It replaces various types of personal information with appropriate placeholder tags to protect privacy while maintaining the overall context of the content.

Inputs

  • Transcript Text: The text content to be processed for PII redaction.

Output

Description

The API returns a JSON response containing the redacted version of the input text as a JSON with key redacted_text. All identified PII elements are replaced with standardized placeholder tags (e.g., [NAME], [EMAIL], etc.).

Example

1{
2  "redacted_text": "Hello, my name is [NAME]. You can reach me at [EMAIL] or [PHONE]. I live at [ADDRESS]."
3}

The API specifically redacts the following types of PII:

  • Names → [NAME]
  • Email addresses → [EMAIL]
  • Phone numbers → [PHONE]
  • Physical addresses → [ADDRESS]
  • Social Security Numbers → [SSN]
  • Credit card numbers → [CREDIT_CARD]
  • Birth dates → [DOB]
  • Other personal identifiers → [PII]