AWS Marketplace
Azure Marketplace
Google Cloud MarketplaceRedact before anything else sees it
Put Philter between speech-to-text and everything downstream, so QA platforms, analytics, and AI features work from redacted transcripts. That can shrink the set of systems in PCI DSS scope.
Contact centers and customer support
PII Redaction for Contact Centers
Call transcripts, chat logs, and tickets are full of names, card numbers, and contact details. Philter redacts them in your own cloud as they come in, before they reach QA tools, analytics, or AI features.
For transcript and chat pipelines
For teams that want it built
What Philterd does with transcripts
Verification codes, not just card numbers
A transcript that captured a card verification code is a common PCI DSS finding, because those codes must not be kept after authorization (requirement 3.3.1). The library’s call transcript policy adds patterns for them, including spoken forms such as “security code.”
Safer agent assist
When agent assist calls an LLM, Philter AI Proxy redacts the prompt before it leaves your network, and can scan the response on the way back.
Which fits your team
Transcript and chat pipelines
Philter
Redacts call transcripts, chat logs, and ticket bodies through an API in your own cloud, as they arrive from your contact center platform.
About Philter →Agent assist and chatbots
Philter AI Proxy
A drop-in proxy that redacts prompts before they reach OpenAI, Anthropic, Gemini, Bedrock, or Ollama, and can scan responses on the way back.
About Philter AI Proxy →A team that wants it built
PII Data Pipelines
We design the redaction, build it inside your cloud, measure it on your own transcripts, and hand you a system you own.
See the engagement →Getting started
- Deploy Philter from your cloud’s marketplace into the account where your transcripts land after speech-to-text.
- Start from the call recording transcripts policy in the policy library, and add patterns for your own account and ticket number formats.
- Measure it on your own transcripts. Have a QA analyst mark a sample of calls, then score the policy against it with Philter Scope. Speech-to-text errors make this step matter more than usual.
- Insert it right after transcription, so every downstream system receives the redacted version.
Ready-to-use redaction policies
Free, ready-to-use policies from the open source policy library. Download and load into your Philter instance.
Contact Center
v1.0.0
Contact Center Call Recording Transcripts
Strip cardholder data and PII from contact-center call transcripts — primarily PAN, CVV, SSN, account numbers — to reduce PCI DSS scope and meet QA privacy requirements.
Finance
v1.0.0
PCI DSS Scope Reduction
Strip cardholder data (PAN, CVV, expiration) from logs, transcripts, and tickets to reduce PCI DSS scope per Requirement 3.4.
General
v2.0.0
CCPA / CPRA Consumer Privacy Redaction
Redact personal information and sensitive personal information as defined by the California Consumer Privacy Act (CCPA/CPRA) from consumer records.
Recent writing on contact center
Redacting PII and PHI from Audio Transcriptions with Philter and Amazon Transcribe
A hands-on guide to redacting PII and PHI from audio transcripts. Run Philter against an Amazon Transcribe transcript and deploy two AWS architectures at scale.
Redaction for Financial Services: PCI DSS, GLBA, and the Real-World Data Pipeline
A practitioner's guide to redacting NPPI and cardholder data in financial workflows, mapping PCI DSS, GLBA, and state requirements to the Philterd toolkit.
Integrating Philter AI Proxy with OpenAI and Anthropic in Production
Pointing a client at the proxy is one line. Where it sits relative to your AI gateway, and what it does with streaming responses, are the parts to get right.