From e9b263cac2eb239d3d443dcd73985bf664bf534f Mon Sep 17 00:00:00 2001
From: karamouche <hugo4ibt@gmail.com>
Date: Wed, 11 Feb 2026 15:29:44 -0500
Subject: [PATCH 1/2] Add pii redaction documentation

---
 chapters/audio-intelligence/index.mdx         |  8 +++
 chapters/audio-intelligence/pii-redaction.mdx | 57 +++++++++++++++++++
 chapters/pre-recorded-stt/features/index.mdx  |  7 +++
 docs.json                                     |  1 +
 4 files changed, 73 insertions(+)
 create mode 100644 chapters/audio-intelligence/pii-redaction.mdx

diff --git a/chapters/audio-intelligence/index.mdx b/chapters/audio-intelligence/index.mdx
index a147226..0c6e055 100644
--- a/chapters/audio-intelligence/index.mdx
+++ b/chapters/audio-intelligence/index.mdx
@@ -32,6 +32,14 @@ Use these capabilities alongside Live or Pre-recorded STT to automate workflows
     Detect and categorize key entities like people, organizations, dates, and more.
   </Card>
 
+  <Card
+    title="PII Redaction"
+    icon="shield-halved"
+    href="/chapters/audio-intelligence/pii-redaction"
+  >
+    Automatically redact names, emails, vehicle IDs, and other PII in pre-recorded transcripts.
+  </Card>
+
   <Card
     title="Chapterization"
     icon="bookmark"
diff --git a/chapters/audio-intelligence/pii-redaction.mdx b/chapters/audio-intelligence/pii-redaction.mdx
new file mode 100644
index 0000000..9589167
--- /dev/null
+++ b/chapters/audio-intelligence/pii-redaction.mdx
@@ -0,0 +1,57 @@
+---
+title: PII Redaction
+description: "Automatically detect and redact personally identifiable information in pre-recorded transcripts"
+---
+
+**PII Redaction** automatically detects and replaces sensitive entities (names, emails, addresses, etc.) in your transcript output. This feature is only available for **pre-recorded** transcription.
+
+## Usage
+
+Add `"pii_redaction": true` to your request to redact all detected PII in the transcript. Sensitive entities will be replaced with markers in the output.
+
+<CodeGroup>
+```json Pre-recorded
+{
+  "audio_url": "YOUR_AUDIO_URL",
+  "pii_redaction": true
+}
+```
+</CodeGroup>
+
+## Optional configuration
+
+You can customize the behavior with `pii_redaction_config`:
+
+<ParamField body="entity_types" type="string[]">
+  Preset or list of PII entity types to redact (e.g. `["GDPR"]`).
+  See [Named Entity Recognition](/chapters/audio-intelligence/named-entity-recognition#supported-entities) for supported entity types.
+</ParamField>
+<ParamField body="processed_text_type" type="string">
+   How to replace detected PII: `"MARKER"` (placeholder labels like `[EMAIL_1]`) or `"MASK"` (masked characters).
+</ParamField>
+
+## Example body
+
+```json
+{
+  "audio_url": "YOUR_AUDIO_URL",
+  "pii_redaction": true,
+  "pii_redaction_config": {
+    "entity_types": ["GDPR"],
+    "processed_text_type": "MARKER"
+  }
+}
+```
+
+## Example output
+
+**Before (raw transcript):**
+
+> Hi, I'm calling about the order for John Smith. Can you confirm the delivery to john.smith@company.com? Yes, John Smith placed it yesterday.
+
+**After (with PII redaction):**
+
+> Hi, I'm calling about the order for [NAME_1]. Can you confirm the delivery to [EMAIL_1]? Yes, [NAME_1] placed it yesterday.
+
+The same entity mentioned multiple times receives the **same marker ID** (e.g. "John Smith" becomes [NAME_1] both times), so you can track references across the transcript while keeping sensitive data redacted. \
+This consistency is also useful for downstream tasks using LLMs, which can reason about entities (e.g. "the person in [NAME_1]") without ever seeing the raw PII.
diff --git a/chapters/pre-recorded-stt/features/index.mdx b/chapters/pre-recorded-stt/features/index.mdx
index 5d75942..88248c9 100644
--- a/chapters/pre-recorded-stt/features/index.mdx
+++ b/chapters/pre-recorded-stt/features/index.mdx
@@ -13,6 +13,13 @@ The core functionality of the Gladia API is its Speech Recognition model, design
   >
     Detect speakers and understand who said what, and when.
   </Card>
+  <Card
+    title="PII Redaction"
+    icon="shield-halved"
+    href="/chapters/audio-intelligence/pii-redaction"
+  >
+    Automatically redact names, emails, vehicle IDs, and other PII in pre-recorded transcripts.
+  </Card>
   <Card
     title="Export subtitles (SRT/VTT)"
     icon="code"
diff --git a/docs.json b/docs.json
index 8fbbdf4..e72af2a 100644
--- a/docs.json
+++ b/docs.json
@@ -95,6 +95,7 @@
               "chapters/audio-intelligence/translation",
               "chapters/audio-intelligence/summarization",
               "chapters/audio-intelligence/named-entity-recognition",
+              "chapters/audio-intelligence/pii-redaction",
               "chapters/audio-intelligence/sentiment-analysis",
               "chapters/audio-intelligence/chapterization",
               "chapters/audio-intelligence/audio-to-llm"

From d3d160d06502ab3e3106e902b91a2aef03f92533 Mon Sep 17 00:00:00 2001
From: karamouche <hugo4ibt@gmail.com>
Date: Wed, 11 Feb 2026 15:42:56 -0500
Subject: [PATCH 2/2] improved explanation on different processed text type for
 PII redaction

---
 chapters/audio-intelligence/pii-redaction.mdx | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/chapters/audio-intelligence/pii-redaction.mdx b/chapters/audio-intelligence/pii-redaction.mdx
index 9589167..16f6187 100644
--- a/chapters/audio-intelligence/pii-redaction.mdx
+++ b/chapters/audio-intelligence/pii-redaction.mdx
@@ -26,8 +26,10 @@ You can customize the behavior with `pii_redaction_config`:
   Preset or list of PII entity types to redact (e.g. `["GDPR"]`).
   See [Named Entity Recognition](/chapters/audio-intelligence/named-entity-recognition#supported-entities) for supported entity types.
 </ParamField>
-<ParamField body="processed_text_type" type="string">
-   How to replace detected PII: `"MARKER"` (placeholder labels like `[EMAIL_1]`) or `"MASK"` (masked characters).
+<ParamField body="processed_text_type" type="enum" default="MARKER">
+   How to replace detected PII: 
+   - **`MARKER`**: Placeholder labels like `[NAME_1]`, `[EMAIL_1]`. Same entity will have same ID.
+   - **`MASK`**: Each character replaced by a mask (e.g. "John Smith" → `#### #####`)
 </ParamField>
 
 ## Example body
@@ -45,11 +47,15 @@ You can customize the behavior with `pii_redaction_config`:
 
 ## Example output
 
-**Before (raw transcript):**
+**Without PII redaction (raw transcript):**
 
 > Hi, I'm calling about the order for John Smith. Can you confirm the delivery to john.smith@company.com? Yes, John Smith placed it yesterday.
 
-**After (with PII redaction):**
+**With PII redaction (`processed_text_type="MASK"`):**
+
+> Hi, I'm calling about the order for #### #####. Can you confirm the delivery to ######################? Yes, #### ##### placed it yesterday.
+
+**With PII redaction (`processed_text_type="MARKER"`):**
 
 > Hi, I'm calling about the order for [NAME_1]. Can you confirm the delivery to [EMAIL_1]? Yes, [NAME_1] placed it yesterday.