Supported Entity Types
Deepgram can detect, format, and redact over 50 unique entity types. The complete inventory of supported entity types is listed in the charts below, divided into four groups: PII (Personally Identifiable Information), PHI (Protected Health Information), PCI (Payment Card Industry), and Other Entities.
When using redaction, you can redact entire groups using redact=pii, redact=phi, or redact=pci. For example, redact=phi will redact all entity types listed in the PHI section below.
Note that some entities, such as name and location, also have subtypes. For instance, location_city is a subtype of location. This means that, in a phrase such as I live in Boston, the location name Boston will be detected as both location and location_city, with the more specific label (in this case, location_city) appearing in the output. Other entity types are groupings of related categories. For example, healthcare_number captures health plan beneficiary numbers and medical record numbers, both of which are outlined as identifiers in the HIPAA Safe Harbor provision. Similarly, numerical_pii covers a broad range of entity types such as MAC addresses and cookie IDs.
While entity types have English names, international variants are also redacted. For example, ssn covers American Social Security Numbers, as well as many equivalent identification numbers used in different regions worldwide, such as the Canadian Social Insurance Number or the German Sozialversicherungsnummer.
Redacting Certain Entities
Deepgram’s redaction functionality supports over 50 unique entity types for both pre-recorded (batch) and streaming requests. Individual entity classes can be redacted by specifying redact=entity_class one or more times.
You can also use redaction groups (pci, pii, phi, numbers) to redact multiple related entity types at once. See the Redaction Groups column in the tables below to see which groups include each entity type.
The numbers group listed in the tables below can be activated with any of the following parameter values: redact=true, redact=numbers, or redact=aggressive_numbers. All three map to the same set of entity types.
Some entity types (cardinal, ordinal, percent) are not included in any redaction group. To redact these, you must specify them individually (e.g. redact=cardinal).
