Video and audio conversion with automatic GPU detection (NVIDIA NVENC, Intel QSV/VAAPI, AMD AMF) and CPU fallback. Browse from root — every NAS volume directly reachable. Currently free.
Video: H.264, H.265, AV1, VP9, MPEG-4. Audio: MP3, AAC, FLAC, Opus, WAV, AC3, Vorbis.
NVIDIA NVENC, Intel Quick Sync (QSV) & VAAPI, AMD AMF — with automatic CPU fallback on errors.
Browse from root, every NAS volume is directly reachable — no need to copy files separately.
Monitor a folder, auto-convert new files, optionally auto-delete after success.
Login via your existing system users (/etc/shadow) — no separate account needed.
Self-hosted, no subscription, runs on a NAS, server or Raspberry Pi.
unzip medienconverter_release.zip -d media-converter && cd media-converter && docker compose up -d --build
http://SERVER-IP:8686 — log in with your NAS/server system user credentials.
Drag & drop a file, choose a target profile (MP4 H.264/H.265, WebM VP9, MP3 …), watch live progress.
Upload multiple files at once or configure a watch folder — queue view shows progress of all jobs.
Set your own ffmpeg parameters for codec, bitrate and extra args — e.g. Plex/Jellyfin-optimized HEVC files.
Automatic switch to software encoding on GPU errors, so no job fails outright.
Docker & Docker Compose required. For GPU acceleration: see FAQ for suitable drivers/group IDs.
unzip medienconverter_release.zip -d media-converter
cd media-converter
docker compose up -d --build
Browser: http://SERVER-IP:8686. Default path: /opt/media-converter/.
Usually incorrect group IDs (check getent group video render on the host) or an unsupported codec profile for the respective GPU generation. Fallback to CPU can be enabled in the settings.
By default 2 GB per file, adjustable via MAX_UPLOAD_MB in the docker-compose.yml (e.g. up to 10 GB). For very large files, the watch folder is recommended instead of upload.
Check the job logs to see whether a GPU encoder (e.g. h264_qsv) is actually being used instead of libx264 (CPU). Even with GPU acceleration, 4K conversions are CPU-intensive during pre-/post-processing.
DRM-protected files, real-time streaming, online URL inputs (not a YouTube downloader), and classic video editor functions like cutting/trimming.
Yes, with CPU fallback (libx264 etc.) — just correspondingly slower, especially with 4K material.
GPU-accelerated batch converter — free for a limited time only.