webui/app.py

@@ -78,6 +78,23 @@ def _run(cmd: list[str], cwd: Path, env: dict | None = None) -> None:
         raise gr.Error(f"Command failed ({cmd[0]}):\n{tail}")
 
 
+def _ensure_wav(path: str) -> str:
+    """AuK's loader (librosa/soundfile) can't read compressed containers like
+    .m4a/.mp3 - only demucs's loader handles those. Normalize once, here, at
+    the point a reference is set, so everything downstream is always wav."""
+    if Path(path).suffix.lower() == ".wav":
+        return path
+    out_path = WORKDIR / f"converted_{int(time.time() * 1000)}.wav"
+    result = subprocess.run(
+        ["ffmpeg", "-nostdin", "-y", "-i", path, "-ar", "44100", "-ac", "1", str(out_path)],
+        capture_output=True, text=True, timeout=120,
+    )
+    if result.returncode != 0:
+        tail = "\n".join(result.stderr.strip().splitlines()[-15:])
+        raise gr.Error(f"Could not convert '{path}' to wav:\n{tail}")
+    return str(out_path)
+
+
 def _ld_library_path_for_venv(venv_dir: Path) -> str:
     site_packages = next(venv_dir.glob(".venv/lib/python*/site-packages"))
     nvidia_dir = site_packages / "nvidia"
@@ -123,6 +140,7 @@ def strip_vocals(input_audio: str):
 def use_as_reference(audio_path: str, history: list):
     if not audio_path:
         raise gr.Error("No audio to use yet.")
+    audio_path = _ensure_wav(audio_path)
     history = [audio_path]
     return audio_path, history, _history_label(history)