Repository navigation
Releases
Releases · chorus-studio/spotify
Beta Release v2.10.1-beta
v2.10.0
🚀 Enhancements
- Add master on/off toggle for the extension (4e635af)
- vocal: Add real-time model-based vocal separation (1450d7f)
- effects: Add live reverb config and EQ band editors (e0e66d9)
- reverb: Split config editor into core and advanced params (9165adb)
- reverb: User-saved config presets + config UX polish (4c1bd66)
- eq: User-saved config presets + parity with reverb, context-aware reset (072fbb7)
🩹 Fixes
- player: Self-heal Chorus controls when Spotify re-renders the now-playing bar (7a90d2b)
- popup: Match on/off power button color to other top-bar icons (59850e2)
- eq: Remap to a full-spectrum 10-band EQ (a4cf761)
- player: Stop volume media keys from seeking via the progress bar (6f2ef60)
- player: Treat missing playback-position as unknown, not 0:00 (0afd768)
- player: Give playback duration the same null discipline as position (73fc52a)
- player: Don't apply old track's snip state during track-change race (43de544)
- player: Don't clamp seeks to the unknown-duration sentinel (e5e966e)
- player: Clear old track's duration and detect song changes flag-free (786e59f)
- snip: Validate snip bounds before saving (795c728)
- spotify: Follow renamed now-playing test ids and album query shape (6d02658)
- skip: Require a track id before blocking from the track list (98cf824)
💅 Refactors
- views: Merge settings and info into one switchable page (98a1186)
- views: Split fx ([eq into separate fx and eq tabs](eq into separate fx and eq tabs))
- reverb: Prune 8 near-duplicate presets with alias fallback (10a2ee8)
📦 Build
- reverb: AOT Faust worklet toolchain (jpverb + zita) (aafcd04)
- deps: Pin svelte-sonner to the published 1.2.1 release (e85ced5)
- vocal: Fetch rt_core_v1 model at build time from a release asset (3d61535)
🏡 Chore
- vocal: Drop dangling vocal-remove.js prototype reference (94fb5d0)
- promo: Ignore Remotion output and keep promo/ out of the sources zip (0e23648)
🤖 CI
- release: Wait for Chrome Web Store upload processing before publishing (f62bccf)
❤️ Contributors
- Cdrani charlesdrani@gmail.com
- Charles (@cdrani)
- A6f00d6
Contributors
cdrani
Assets 4
Model asset: rt_core_v1 (vocal separation)
Hosts the real-time vocal-separation model used by the Chrome build of Chorus.
This is not an extension release. Extension releases are the v* tags. This release only exists so that clean checkouts (CI) can fetch the model at build time via scripts/fetch-model.ts, since the ~50 MiB file is too large to live in the repository.
- File:
rt_core_v1.bin(small-frame MDX-Net from Ultimate Vocal Remover, MIT — seesrc/public/models/rt_core_v1.LICENSE) - Size: 52,786,724 bytes
- SHA-256:
8303cb0e37fc3d301130745502c9a82a4f650ee129897bdb1525914bda7af72e
The fetch script verifies both the size and the checksum before using the file.
v2.9.0
🚀 Enhancements
- background: Add onboarding page on install and uninstall feedback URL (0e4ae4e)
🩹 Fixes
- volume: Preserve muted state across page loads and track skips (6575017)
- audio: Guard effect updates until audio chain is initialized (0819c03)
- pip: Resolve audio cutoff and seek in Picture-in-Picture (ca3b7fc)
- audio: Reduce impulse reverb intensity and restore mono input for digital reverb (19383f4)
- types: Resolve TypeScript narrowing error in MediaElement play handler (5fd2d3b)
❤️ Contributors
- Cdrani (@cdrani)
Contributors
cdrani