I-IAL9000/shrinkerr v0.9.80

on GitHub

Changed

• Audio language detection loads the Whisper model once instead of per clip. A single persistent, killable worker now serves all clips (released after idle, respawned on timeout/crash/model change), so detection is much faster over large libraries and larger models become practical. Killability (v0.9.21) is unchanged.

Docker images

Pinned to this release:

docker pull ghcr.io/i-ial9000/shrinkerr:v0.9.80         # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:v0.9.80-nvenc   # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:v0.9.80-edge    # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:v0.9.80-edge-nvenc

Floating tags (:latest, :edge, :nvenc, :edge-nvenc) track the most recent release of each lineage.