Audio Anonymization

The Audio anonymization feature allows you to listen to sample audio scenarios and see how anonymization is applied to sensitive data. Currently, real-time anonymization testing for audio is not available, but you can explore different sample scenarios in the demo.

How to Use the Audio Anonymization Feature

1. Select Sample Scenario:
   Choose from various pre-defined audio scenarios available in the dropdown. Example scenarios include:

   • Healthcare Call: Anonymize sensitive information in a healthcare conversation.
   • Customer Support: Anonymize sensitive customer data during support calls.
   • Legal Consultation: Anonymize sensitive information shared during legal consultations.

2. Listen to the Audio:
   After selecting a scenario, you can play the sample audio. The audio player lets you listen to the scenario and understand how the anonymization process works.

3. Convert to Text:
   Once you've listened to the sample audio, you can click the Convert to Text button to convert the audio into text format. This allows you to view the anonymized text data that corresponds to the audio.

4. Select Anonymization Method:
   You can choose an anonymization method to apply to the text after conversion. The available methods are:

   • Replace: Replace detected entities with a placeholder (e.g., “******”).
   • Redact: Completely hide detected sensitive information (e.g., “REDACTED”).
   • Mask: Mask the detected information by replacing characters with a specific symbol (e.g., “*****”).
   • Tokenization: Replace sensitive information with a token, which is a randomly generated identifier that can be later mapped back to the original data.
   • Encryption: Encrypt the detected sensitive data so it is stored in a secure format, only accessible with a decryption key.
   • Pseudonymization: Replace sensitive data with pseudonyms, ensuring that the data cannot be traced back to an individual without additional information.
   • Hash: Hash the detected data, turning it into a fixed-size string that represents the original data, but cannot be reverted back to the original value.

5. Allow List:
   The Allow List feature lets you specify entities that should not be anonymized, even if they are detected. Enter any terms or values you wish to keep intact in the Allow List.

6. Deny List:
   The Deny List ensures that certain entities are always anonymized, regardless of whether they are detected automatically. Add items to the Deny List to force their anonymization.

7. Select Entities to Detect:
   You can choose which types of sensitive information you want the tool to detect and anonymize. If you want to anonymize all sensitive entities, leave No selection checked. If you only want to anonymize specific types of data (e.g., credit cards, email addresses), select the appropriate entities.

   Available entities to detect include:

   • Credit Card
   • Crypto
   • Payment Information
   • Date/Time
   • Email Address
   • IBAN Code
   • IP Address
   • National Registration Number

8. Apply Anonymization:
   Once everything is configured, click the Apply Anonymization button to process the text and anonymize the detected entities according to your selected anonymization method.

---

While real-time audio anonymization isn't available yet, this demo will give you a great idea of how the tool processes sensitive audio data for anonymization and privacy compliance.