@ -232,7 +243,7 @@ Open SoulSync at `http://localhost:8008` (or your server's IP) and go to Setting
**Paths:**
- **Download Path**: This should be the container path where slskd's downloads are mapped. If you followed the volume mapping above, that's `/app/downloads`
- **Transfer Path**: This is where SoulSync puts your organized/renamed music. The default is `/app/Transfer` - just make sure you have a volume mapping for it so the files persist and your media server can see them
- **Staging Path**: Optional — for importing existing audio files. Map a host folder to `/app/Staging` if you want to use the import feature
- **Staging Path**: Optional — for the Import page. Map a host folder to `/app/Staging`, drop audio files in, and use the Import page to match them to albums/tracks with full metadata processing
**Media Server** (optional):
- **Plex/Jellyfin/Navidrome URL**: Use your machine's actual IP address, not `localhost`. For example `http://192.168.1.100:32400` for Plex. Same reason as slskd - `localhost` inside Docker means the container, not your host machine.
@ -313,6 +324,7 @@ If you're using the Unraid template from Community Applications:
@ -337,7 +349,7 @@ If you're using the Unraid template from Community Applications:
**Core Components**:
- **Matching Engine** — version-aware fuzzy matching with multi-strategy query generation and source-specific confidence weighting
- **Download Orchestrator** — routes between Soulseek and YouTube with hybrid fallback, batch processing with concurrent workers, automatic retry on failure/timeout
- **Download Orchestrator** — routes between Soulseek, Tidal, and YouTube with hybrid fallback, batch processing with concurrent workers, automatic retry on failure/timeout
- **Discovery System** — custom algorithms for personalized playlists, seasonal content, and similar artist exploration
- **Metadata Pipeline** — dual-source (Spotify/iTunes) with MusicBrainz enrichment, AcoustID verification, LRC lyrics, album art embedding via mutagen
- **Database Update Worker** — incremental sync from media servers with smart early-stopping (Jellyfin fast-path: ~2 API calls vs thousands)