Added
- Language detection reads the track title. A language named in a track's title ("Traditional Chinese", "Romanian", "English SDH") is now used directly, tried before content/OCR detection. Fixes
undforced/SDH subtitles that carry too little text for langdetect but name the language in their title. Applies to audio and subtitle tracks.
Docker images
Pinned to this release:
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.10 # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.10-nvenc # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.10-edge # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:0.9.10-edge-nvencFloating tags (:latest, :edge, :nvenc, :edge-nvenc) track the most recent release of each lineage.