Self-host subarr
v2.7.24Subtitle coordination layer for the *arr stack
- Stars
- 147
- CPU
- 0.25 cores (reserved)
- RAM
- 256 MB (reserved)
- Port
- 9922 (HTTP web UI and API)
- Requires
- Bazarr, Radarr, Sonarr, Subgen
ghcr.io/coaxk/subarr
What is subarr?
Subarr is a coordination service that sits alongside Bazarr in the *arr media automation stack. It analyzes your media library to determine which subtitles are actually missing, verifies audio languages by sampling multiple chunks of each file with Whisper, and schedules transcription jobs to subgen (a patched Whisper worker) with backpressure and priority signals from Tautulli. It provides a web UI with dashboard, coverage gap analysis, queue management, library browser, audio-language review with an audio player, a Whisper tuning lab for per-language model optimization, and provenance tracking of which provider delivered each subtitle. It integrates with Sonarr, Radarr, Bazarr, Plex, Jellyfin, Tautulli, and optionally Ollama for AI enrichment. Runs as a non-root Docker container with SQLite storage on a persistent volume. Authentication is enabled by default with a single admin account, API keys, and reverse-proxy support.
Should you self-host subarr?
subarr scores 48/100 on selfhost.directory (how scoring works) and ranks #79 of 142 in Media Management. It has 147 GitHub stars.
It shipped 10 stable releases in the past year. The latest, v2.7.24, came out on 22 October 2026. New versions land roughly every 2 days.
Setup is advanced: plan for several services and some ongoing maintenance. An official Docker image is available. It is MIT-licensed and written mostly in Python.
Its closest alternative on selfhost.directory is Sonarr (81/100, 17 471 stars). It scores higher overall, so compare the two before you pick. See subarr vs Sonarr side by side.
Features
- Audio-language review queue
- Auto-queue rules with score thresholds
- Auto-re-timing of generated subtitles
- Calibrated audio-language detection
- Default audio track mismatch detection
- Forced subtitle segment generationExperimental, off by default
- Image-only subtitle filling
- Multi-instance stack support
- Multi-library support
- Post-transcription quality review (Aftercare)
- Provenance ledger
- Scheduled transcription with backpressure
- Subtitle coverage analysis
- Whisper Tuning Lab
subarr README
More in Audio Processing
SubgenSelf-hosted subtitle generator using OpenAI Whisper. Integrates with Bazarr, Plex, Jellyfin, Emby, Tautulli. Runs on CPU or NVIDIA GPU via Docker.
tele-vdo-riderSelf-hosted Telegram bot that converts videos from YouTube and other sites to MP3 using youtube-dl and ffmpeg, with automatic file splitting for Telegram's size limits.
transcriberSelf-hosted, open-source web app for transcribing audio/video with Whisper. Supports YouTube, local files, real-time progress, dark mode.
txtifySelf-hosted web app for transcribing YouTube videos and media files using Whisper AI models, with translation and multiple export formats.
WhishperSelf-hosted open-source audio transcription and subtitling suite running locally with a web UI, using FasterWhisper and LibreTranslate.
whisper-api-serverA self-hosted, open-source transcription server that implements the OpenAI Whisper API locally using faster-whisper for efficient speech-to-text conversion.
whisper-transcriber-botSelf-hosted Telegram bot that converts voice messages and audio files into text using a local AI model, running on CPU without needing a GPU or cloud APIs.
yoinkyoink is a self-hosted, open-source media tool that downloads from 1000+ sites, converts formats, compresses for Discord, and offers transcription — no ads or tracking.
audiobook-studioSelf-hosted open-source app for producing audiobooks locally using XTTS voice cloning, with optional Voxtral cloud voices. Multi-voice, segment repair, chapter assembly
audiodeckAudioDeck is a self-hosted, open-source web app that generates browser spectrograms to verify audio quality of MP3, FLAC, and WAV files with a lightweight Go backend.
AudioMuse-AIOpen-source, self-hosted sonic-analysis tool that builds playlists, a 2D music map and semantic search for Navidrome, Jellyfin, Emby, Plex, LMS and Lyrion libraries.
by2kbby2kb is a self-hosted CLI turning Bilibili/YouTube URLs into Markdown transcripts, abstracts, and study notes via local Whisper or cloud ASR.
0 comments