groupultra/telegram-search
Telegram Search
Telegram Search is a self-hosted platform designed to export, index, and archive personal or group message history. It functions as a private search engine that transforms scattered communication logs and media assets into a searchable knowledge library, allowing users to maintain full control over their data through containerized infrastructure.
The platform distinguishes itself by utilizing vector-based semantic indexing to enable fuzzy retrieval across historical datasets. It incorporates an optical character recognition pipeline to extract text from images and media files, ensuring that visual content is as discoverable as text-based messages. Users can navigate directly from search results back to the original source within the messaging application using protocol-based deep links.
Beyond basic search, the system integrates generative artificial intelligence to provide context-aware summaries and answers based on stored chat logs. This retrieval-augmented generation capability allows for intelligent analysis of historical threads, while automated media archiving offloads heavy assets to external storage to maintain a lightweight local database. The entire stack is deployed via containerized configurations to simplify the management of the interface, database, and storage services.
Features
- Telegram Chat Importers - Exports and preserves personal or group message history into a private database for long-term storage and offline access.
- Retrieval Augmented Generation - Combines stored chat archives with large language models to provide context-aware answers and summaries based on private user data.
- Semantic Vector Search - Performs fuzzy and vector-based searches across text and images to quickly locate specific information within archived communication history.
- Vector Indexing - Converts message content into high-dimensional numerical embeddings to enable fast similarity searches across large historical datasets.
- Chat Analysis - Analyzes historical and unread message threads to provide intelligent summaries and context-aware answers to user queries.
- Local Chat Archiving - Provides a containerized infrastructure for backing up and organizing chat media and message history into local or remote storage.
- Retrieval Augmented Generation Platforms - Analyzes stored chat logs to provide automated summaries and context-aware answers to user queries.
- Vector Databases and Search - Utilizes embeddings to enable fuzzy and semantic retrieval of historical text and image content.
- Personal Knowledge Management - Indexes chat content and media assets to turn scattered conversations into a searchable library of information and shared resources.
- Chat History Synchronization - Synchronizes and stores message data into local or remote databases, processing content for vector-based search to ensure conversations remain accessible.
- Chat Query Engines - Enables retrieval-augmented generation to answer questions using chat history as context, helping users manage communication more effectively.
- Optical Character Recognition Indexers - Extracts embedded text from images and media files during the ingestion process to make visual content searchable by keyword.
- Media Library Indexers - Automatically indexes shared media and links through text recognition and image description, while backing up files to secure storage.
- Object Storage Persistence - Offloads heavy media assets to external storage services to maintain a lightweight local database while ensuring long-term data preservation.
- Deep Linking Utilities - Provides direct protocol-based links from search results to the original message source for context verification.
- Fuzzy File Content Searches - Performs multi-language fuzzy and semantic searches across text and images using vector embeddings to locate specific historical conversations.
- Containerized Service Deployment - Supports containerized deployment of the interface, database, and storage services using a single configuration file to simplify private infrastructure hosting.
- Multi-Service Container Orchestration - Packages the application, database, and storage components into isolated environments to simplify self-hosting and infrastructure management.
- Self-Hosted Infrastructure - Deploys and manages private search and storage services using containerized configurations to maintain full control over data.
