feat: implement TTS, Document processing, and Memory Service /facts API

- TTS: xtts-v2 integration with voice cloning support
- Document: docling integration for PDF/DOCX/PPTX processing
- Memory Service: added /facts/upsert, /facts/{key}, /facts endpoints
- Added required dependencies (TTS, docling)

services/chandra-service/README.md

@@ -0,0 +1,61 @@
+# Chandra Document Processing Service
+
+Wrapper service for Datalab Chandra OCR model for document and table processing.
+
+## Features
+
+- Document OCR with structure preservation
+- Table extraction with formatting
+- Handwriting recognition
+- Form processing
+- Output formats: Markdown, HTML, JSON
+
+## API Endpoints
+
+### Health Check
+```
+GET /health
+```
+
+### Process Document
+```
+POST /process
+```
+
+**Request:**
+- `file`: Uploaded file (PDF, image)
+- `doc_url`: URL to document
+- `doc_base64`: Base64 encoded document
+- `output_format`: markdown, html, or json
+- `accurate_mode`: true/false
+
+**Response:**
+```json
+{
+  "success": true,
+  "output_format": "markdown",
+  "result": {
+    "markdown": "...",
+    "metadata": {...}
+  }
+}
+```
+
+### List Models
+```
+GET /models
+```
+
+## Configuration
+
+Environment variables:
+- `CHANDRA_API_URL`: URL to Chandra inference service (default: `http://chandra-inference:8000`)
+- `CHANDRA_LICENSE_KEY`: Datalab license key (if required)
+- `CHANDRA_MODEL`: Model name (chandra-small or chandra)
+
+## Integration
+
+This service integrates with:
+- Router (`OCR_URL` and `CHANDRA_URL`)
+- Gateway (`doc_service.py`)
+- Memory Service (for storing processed documents)