dj sets + lyrics: a mix that names its own tracks, and words pulled off the separated voice

src-tauri/src/config.rs

@@ -139,6 +139,7 @@ pub struct AppConfig {
     /// every track plays at roughly the same perceived level. None → true.
     #[serde(default)]
     pub volume_normalization: Option<bool>,
+    pub word_timing: Option<bool>,
 }
 
 /// Default longest edge for embedded covers when the user hasn't chosen.
@@ -180,6 +181,14 @@ pub fn volume_normalization(cfg: &AppConfig) -> bool {
     cfg.volume_normalization.unwrap_or(true)
 }
 
+/// Word-by-word lyric timing (the karaoke fill) — EXPERIMENTAL, off by
+/// default: generated word spans still drift on dense or echo-heavy mixes,
+/// and the solid line-level sync must be what users live with until the
+/// feature earns its keep (Kyle, 2026-08-19).
+pub fn word_timing(cfg: &AppConfig) -> bool {
+    cfg.word_timing.unwrap_or(false)
+}
+
 /// One-time migration after the identifier rename: the app-data dir moved
 /// from `com.trifecta.soulseek-plus` to the new identifier's folder. Rename
 /// the old dir into place so the library DB, config and stream cache