steipete/summarize v0.6.0 on GitHub

Podcasts (full episodes)
- Support Apple Podcasts episode URLs via iTunes Lookup + enclosure transcription (avoids slow/blocked HTML).
- Support Spotify episode URLs via the embed page (/embed/episode/...) to avoid recaptcha; fall back to iTunes RSS when embed audio is DRM/missing.
- Prefer local whisper.cpp when installed + model available (no API keys required for transcription).
- Whisper transcription works for any media URL (audio/video containers), not just YouTube.
Language
- Add --language/--lang (default: auto, match source language).
- Add config support via output.language (legacy language still supported).
Progress UI
- Add two-phase progress for podcasts: media download + Whisper transcription progress.
- Show transcript phases (YouTube caption/Apify/yt-dlp), provider + model, and media size/duration.

Transcription
- Add lenient ffmpeg transcode fallback for local Whisper when strict decode fails (e.g. Spotify AAC).
Models
- Add zai/... model alias with Z.AI base URL + chat completions by default.
- Add OPENAI_USE_CHAT_COMPLETIONS + openai.useChatCompletions config toggle.
Metrics / output
- --metrics on|detailed: finish line includes compact transcript stats (… words, …) + media duration (when available); --metrics detailed: also prints input/transcript sizes + transcript source/provider/cache; hides calls=1.
- Smarter duration formatting (1h 13m 4s, 44s) and rounded transfer rates.
- Make Markdown links terminal-clickable by materializing URLs.
- --metrics on|detailed renders a single finish line with a compact transcript block (… words, …) before the model.
Cost
- Include OpenAI Whisper transcription estimate (duration-based) in the finish line total (txcost=…); configurable via openai.whisperUsdPerMinute.

Add JSON-LD graph extraction coverage.
Extend live podcast-host coverage (Podchaser, Spreaker, Buzzsprout).
Raise global branch coverage threshold to 75% and add regression coverage for podcast/language/progress paths.