Skip to content

The open-source AI voice studio. Clone, dictate, create.

fix: torch.from_numpy crash with numpy 2.x in frozen binary (#361)

aimaaaimaaa383ff68637b
diff --git a/backend/build_binary.py b/backend/build_binary.py
index 43ad0719..84c019ab 100644
--- a/backend/build_binary.py
+++ b/backend/build_binary.py
@@ -52,6 +52,16 @@ def build_server(cuda=False):
if platform.system() == "Windows":
args.append("--noconsole")
+ # numpy 2.x / torch ABI mismatch fix: install memmove fallback for
+ # torch.from_numpy() before the app starts. Runtime hooks run after
+ # FrozenImporter is registered so frozen torch/numpy are importable.
+ args.extend(
+ [
+ "--runtime-hook",
+ str(backend_dir / "pyi_rth_numpy_compat.py"),
+ ]
+ )
+
# Add local qwen_tts path if specified (for editable installs)
qwen_tts_path = os.getenv("QWEN_TTS_PATH")
if qwen_tts_path and Path(qwen_tts_path).exists():
diff --git a/backend/pyi_rth_numpy_compat.py b/backend/pyi_rth_numpy_compat.py
new file mode 100644
index 00000000..2e3b5468
--- /dev/null
+++ b/backend/pyi_rth_numpy_compat.py
@@ -0,0 +1,80 @@
+"""
+PyInstaller runtime hook: numpy 2.x / torch ABI mismatch fix.
+
+Problem
+-------
+torch is compiled against numpy 1.x headers. numpy 2.x changed the version
+number returned by PyArray_GetNDArrayCVersion() (0x01000009 → 0x02000000),
+so torch's is_numpy_available() returns False and every torch.from_numpy()
+call raises:
+
+ RuntimeError: Numpy is not available
+
+This surfaces as:
+
+ ValueError: Unable to create tensor, you should probably activate
+ padding with 'padding=True'
+
+during TTS generation (EncodecFeatureExtractor → BatchFeature.convert_to_tensors).
+
+Fix
+---
+Runtime hooks execute after PyInstaller's FrozenImporter is registered, so
+frozen torch/numpy are importable here. We start a background thread that
+waits for torch to finish loading then wraps torch.from_numpy with a ctypes
+memmove fallback that bypasses the C-level numpy ABI check entirely.
+
+This approach works with any numpy version and is safer than binary-patching
+libtorch_python.dylib (which risks PyArray_Descr struct layout mismatches).
+"""
+
+import sys
+import threading
+
+
+def _patch_torch_from_numpy():
+ import time
+
+ for _ in range(7200): # poll up to 360 s at 50 ms intervals
+ time.sleep(0.05)
+ torch = sys.modules.get("torch")
+ if torch is None or not hasattr(torch, "from_numpy"):
+ continue
+ if getattr(torch, "_vb_from_numpy_patched", False):
+ return
+ try:
+ import ctypes
+ import numpy as np
+
+ _orig = torch.from_numpy
+
+ def _safe_from_numpy(arr, _orig=_orig, _c=ctypes, _np=np, _t=torch):
+ try:
+ return _orig(arr)
+ except RuntimeError:
+ a = _np.ascontiguousarray(arr)
+ dtype_map = {
+ "float32": _t.float32,
+ "float64": _t.float64,
+ "int32": _t.int32,
+ "int64": _t.int64,
+ "int16": _t.int16,
+ "int8": _t.int8,
+ "uint8": _t.uint8,
+ "bool": _t.bool,
+ }
+ out = _t.empty(
+ list(a.shape),
+ dtype=dtype_map.get(str(a.dtype), _t.float32),
+ )
+ _c.memmove(out.data_ptr(), a.ctypes.data, a.nbytes)
+ return out
+
+ torch.from_numpy = _safe_from_numpy
+ torch._vb_from_numpy_patched = True
+ except Exception:
+ pass
+ return
+
+
+threading.Thread(target=_patch_torch_from_numpy, daemon=True).start()
diff --git a/backend/requirements.txt b/backend/requirements.txt
index e916b1d2..c8aa1877 100644
--- a/backend/requirements.txt
+++ b/backend/requirements.txt
@@ -50,7 +50,7 @@ en_core_web_sm @ https://github.com/explosion/spacy-models/releases/download/en_
# Audio processing
librosa>=0.10.0
soundfile>=0.12.0
-numpy>=1.24.0
+numpy>=1.24.0,<2.0
numba>=0.60.0,<0.61.0
pedalboard>=0.9.0