Identify and Remove Personally Identifiable Information from Audio, Images, and Video

Your Multi-Modal PII Tool

Remove PII from audio recordings, images, and video files to protect your customers’ sensitive data and comply with global privacy regulations like the GDPR, CPRA, HIPAA, and more.

Private AI can be integrated into any cloud infrastructure or database to identify and redact PII to help with critical data discovery and data classification projects. We deploy as a Docker container so your data never leaves your environment and is never shared with an external party (including Private AI). 

Built by experts from:

Built by experts from:


Private AI removes PII from audio recordings with customizable bleeping for basic data protection. For compliance with privacy regulations, like the GDPR, Private AI also offers voice and tone distortions to effectively de-identify audio recordings.  

Compatible with 10+ audio formats, including WAV and MP3. Contact us for documentation and more information.

Image and Video

Private AI blurs faces and personally identifiable text (eg. licence plates, credit card details, addresses, and more) from images and video to protect your customer data and comply with privacy regulations. View the full list of entities we capture.  

Compatible with 10+ image and video formats. Contact us for documentation and more information. 



Private AI can also find PII in a range of different document types, such as PDF and DOCX. In addition to scanning text fields, Private AI also scans embedded graphics for complete coverage. After PII has been identified, a copy of the document with all PII blurred can be created.

Contact us for documentation and more information.

Get Started Today


Testé sur un ensemble de données composé de données conversationnelles désordonnées contenant des informations de santé sensibles. Téléchargez notre livre blanc pour plus de détails, ainsi que nos performances en termes d’exactitude et de score F1, ou contactez-nous pour obtenir une copie du code d’évaluation.

99.5%+ Accuracy

Number quoted is the number of PII words missed as a fraction of total number of words. Computed on a 268 thousand word internal test dataset, comprising data from over 50 different sources, including web scrapes, emails and ASR transcripts.

Please contact us for a copy of the code used to compute these metrics, try it yourself here, or download our whitepaper.