Skip to main content

OpenTranscribe

AI-Powered Transcription and Media Analysis Platform

OpenTranscribe Complete Workflow

Complete workflow: Login → Upload → Process → Transcribe → Speaker Identification → AI Tags & Collections

Self-hosted, open-source AI transcription with speaker identification, AI summarization, and powerful search. Privacy-first, production-ready, and 70x realtime speed.

curl -fsSL https://raw.githubusercontent.com/davidamacey/OpenTranscribe/master/setup-opentranscribe.sh | bash
70x
Realtime Speed
50+
Languages
100%
Local & Private
AGPL-3.0
Open Source

Why OpenTranscribe?

🎧

High-Accuracy Transcription

Powered by WhisperX with word-level timestamps and 70x realtime speed on GPU. Supports 50+ languages with automatic English translation.

👥

Smart Speaker Detection

Automatic speaker diarization with PyAnnote.audio. Cross-video speaker recognition with voice fingerprinting and AI-powered identification.

🤖

AI-Powered Insights

Generate BLUF summaries, extract action items, and analyze speaker patterns using multiple LLM providers (OpenAI, Claude, vLLM, Ollama).

🔍

Hybrid Search

Lightning-fast full-text and semantic search with OpenSearch 3.3.1. 9.5x faster vector search and 25% faster queries.

🔒

Privacy-First

Self-hosted solution where all data stays on your infrastructure. Works completely offline. Perfect for sensitive content.

Production-Ready

Docker-based deployment with GPU acceleration, multi-worker architecture, and comprehensive monitoring. Scales from laptop to datacenter.

Ready to Get Started?

Install OpenTranscribe in less than 5 minutes with our one-line installer.

Install Now →