I-IAL9000/shrinkerr v0.8.2

on GitHub

Added

• Charset detection for legacy non-Unicode subtitles. Subtitles in GB2312/GBK/Big5 (Chinese), Shift-JIS (Japanese), EUC-KR (Korean), or Windows-1252/Latin-1 were being blind-decoded as Latin-1 — turning CJK double-byte text into mojibake that language detection couldn't read, leaving the track und. Now uses charset-normalizer to detect the real encoding before decoding, so these subs detect correctly (e.g. a GB2312 Chinese sub → chi). langdetect's regional codes (zh-cn/zh-tw) are normalized to chi.

Changed

• Conversion shows a "Detecting language…" status step while faster-whisper runs on und audio tracks, instead of a silent pause before the encode.
• The "Auto-detect languages" setting now appears in both the audio and subtitle sections (one shared setting, shown in both places).

Docker images

Pinned to this release:

docker pull ghcr.io/i-ial9000/shrinkerr:0.8.2         # portable (amd64 + arm64, CPU)
docker pull ghcr.io/i-ial9000/shrinkerr:0.8.2-nvenc   # x86_64 + NVIDIA GPU (NVENC)
docker pull ghcr.io/i-ial9000/shrinkerr:0.8.2-edge    # as above but ffmpeg master
docker pull ghcr.io/i-ial9000/shrinkerr:0.8.2-edge-nvenc

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