got-feedBack/feedBack
1
0
Fork 0
mirror of https://github.com/got-feedBack/feedBack.git synced 2026-08-11 03:09:57 +00:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Repoint dead slopsmith URLs -> got-feedback

Browse Source 
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
byrongamatos
2026-06-17 11:02:04 +02:00
co-authored by Claude Opus 4.8
parent 97a1e5fbf5
commit edf8f46866
19 changed files with 59 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/vocal_pitch.py
+2 -2
View File
@@ -5,7 +5,7 @@ isolated vocals + per-syllable lyric timing (both produced by the
WhisperX fallback or shipped in the source archive), the /pitch endpoint
runs CREPE over the vocals stem and returns one MIDI note per supplied
timing token. The result lands in `<sloppak>/vocal_pitch.json` in the
shape the byrongamatos/slopsmith-plugin-lyrics-karaoke renderer
shape the got-feedback/feedback-plugin-lyrics-karaoke renderer
already consumes:
{"version": 1, "notes": [{"t": float, "d": float, "midi": int}, ...]}
@@ -23,7 +23,7 @@ runs locally. Adding a local CREPE path here would mean pulling
`crepe` + `tensorflow` as plugin deps (~500 MB+ on top of the
existing torch/demucs/whisperx). Deferred until users hit the gap.
If you need a local fallback today, install
`byrongamatos/slopsmith-plugin-lyrics-karaoke` and let its local
`got-feedback/feedback-plugin-lyrics-karaoke` and let its local
pYIN run when the server isn't reachable.
Cache key parity with stem_separation / lyric_transcription