Redaction
Redaction removes sensitive information from your transcripts.
redact
boolean. Default: false
Deepgram’s Redaction feature redacts sensitive information.
Enable Feature
To enable redaction, use the following parameter in the query string when you call Deepgram’s /listen
endpoint:
redact=OPTION
Hosted
Redaction has the following options available for those using Deepgram's hosted endpoint (api.deepgram.com
).
Pre-Recorded
When submitting pre-recorded audio to Deepgram's hosted endpoint, you may select the types of entities you wish to redact from over 50 supported entity types. This powerful functionality allows you total control over what is redacted in your transcript.
In addition to specifying individual types of entities for redaction, Deepgram provides the following options to redact common groups of entities:
pci
: Redacts credit card information, including credit card number, expiration date, and CVVpii
: Redacts personally identifiable information, including names and locationsnumbers
(ortrue)
: Aggressively redacts strings of numbers
Multiple types of entities can be redacted with the syntax redact=option_1&redact=option_2
. For example, redact=email_address&redact=pci
.
To transcribe audio from a file on your computer, run the following cURL command in a terminal or your favorite API client.
curl \
--request POST \
--header 'Authorization: Token YOUR_DEEPGRAM_API_KEY' \
--header 'Content-Type: audio/wav' \
--data-binary @youraudio.wav \
--url 'https://api.deepgram.com/v1/listen?redact=OPTION'
Replace
YOUR_DEEPGRAM_API_KEY
with your Deepgram API Key.
Streaming
Live streamed redaction is not currently available when using smart formatting on our Nova or enhanced tier models.
This feature is available for English only (all available regions).
When live-streaming audio to Deepgram's hosted endpoint, redaction options include:
pci
: Redacts sensitive credit card information, including credit card number, expiration date, and CVVnumbers
(ortrue)
: Aggressively redacts strings of numeralsssn
: Redacts social security numbers
Multiple redaction values can be sent: redact=pci&redact=numbers
On-Prem Deployments
Deepgram's on-prem deployments offer the following set of redaction options. The available options do not differ between pre-recorded and live-streamed audio.
Possible options include:
pci
: Redacts sensitive credit card information, including credit card number, expiration date, and CVVnumbers
(ortrue)
: Aggressively redacts strings of numeralsssn
: Redacts social security numbers
Multiple redaction values can be sent: redact=pci&redact=numbers
Live streamed redaction is not currently available when using smart formatting on our Nova or enhanced tier models.
This feature is available for English only (all available regions).
Results
Hosted
Redaction has the following output for those using Deepgram's hosted endpoint (api.deepgram.com
).
Pre-Recorded
For pre-recorded audio, redaction replaces redacted content with the type of entity redacted and the number of times that entity has been detected in the transcript. For example, if you choose to redact social security numbers, the phrase "My social security number is five five five two two one one one one and his is six six six two two one three three three" would appear in your transcript as "My social security number is [SSN_1] and his is [SSN_2]".
Example with redact=pci&redact=pii
:
Truth | Before redact | After redact |
---|---|---|
my credit card number is four four four four nine nine nine nine three three three three two two two two with an expiration date of one twenty three and the cvv code is one one one i live at one two three main street dallas texas seven five two zero one my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six | my credit card number is four four four four nine nine nine nine three three three three two two two two with an expiration date of one twenty three and the cvv code is one one one i live at one two three main street dallas texas seven five two zero one my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six | my credit card number is [CREDIT_CARD_1] with an expiration date of [CREDIT_CARD_EXPIRATION_1] and the cv code is [CVV_1] i live at [LOCATION_ADDRESS_1] my phone number is [PHONE_NUMBER_1] my date of birth is [DOB_1] |
Live Streaming Audio
For streaming audio, redaction replaces redacted content with asterisks (*).
Example with redact=pci
:
Truth | Before redact | After redact |
---|---|---|
my credit card number is four four four four nine nine nine nine three three three three two two two two with an expiration date of one twenty three and the cvv code is one one one i live at one two three main street dallas texas seven five two zero one my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six | my credit card number is four four four four nine nine nine nine three three three three two two two two with an expiration date of one twenty three and the cvv code is one one one i live at one two three main street dallas texas seven five two zero one my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six | my credit card number is * with an expiration date of * and the cvv code is * i live at * main street dallas texas * my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six |
On-Prem Deployments
Redaction replaces redacted content with asterisks (*).
Example with redact=pci
:
Truth | Before redact | After redact |
---|---|---|
my credit card number is four four four four nine nine nine nine three three three three two two two two with an expiration date of one twenty three and the cvv code is one one one i live at one two three main street dallas texas seven five two zero one my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six | my credit card number is four four four four nine nine nine nine three three three three two two two two with an expiration date of one twenty three and the cvv code is one one one i live at one two three main street dallas texas seven five two zero one my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six | my credit card number is * with an expiration date of * and the cvv code is * i live at * main street dallas texas * my phone number is five five five two one two nine three three three my date of birth is july twelfth nineteen seventy six |
Use Cases
Some examples of use cases for redaction include:
- Customers who handle sensitive customer information, including social security numbers or credit card information, and need to ensure this data isn’t stored for security and compliance purposes.
By default, Deepgram applies its general AI model, which is a good, general purpose model for everyday situations. To learn more about the customization possible with Deepgram's API, check out the Deepgram API Reference.
Updated 8 days ago