Check if libsndfile is available by importing soundfile (#1718)

`torchaudio` treats `soundfile` as an optional dependency, and it assumes either `soundfile` is properly installed or it is not installed at all. However, there is a third state where `soundfile` is installed but the backing library `libsndfile` is not installed, and in this case, `import torchaudio` fails.

This commit resolves it by further checking if `soundfile` is importable. 

See also: https://github.com/pytorch/audio/issues/1687

torchaudio/_internal/module_utils.py

@@ -77,6 +77,37 @@ def requires_kaldi():
     return decorator
 
 
+def _check_soundfile_importable():
+    if not is_module_available('soundfile'):
+        return False
+    try:
+        import soundfile    # noqa: F401
+        return True
+    except Exception:
+        warnings.warn("Failed to import soundfile. 'soundfile' backend is not available.")
+        return False
+
+
+_is_soundfile_importable = _check_soundfile_importable()
+
+
+def is_soundfile_available():
+    return _is_soundfile_importable
+
+
+def requires_soundfile():
+    if is_soundfile_available():
+        def decorator(func):
+            return func
+    else:
+        def decorator(func):
+            @wraps(func)
+            def wrapped(*args, **kwargs):
+                raise RuntimeError(f'{func.__module__}.{func.__name__} requires soundfile')
+            return wrapped
+    return decorator
+
+
 def is_sox_available():
     return is_module_available('torchaudio._torchaudio') and torch.ops.torchaudio.is_sox_available()
 

torchaudio/backend/soundfile_backend.py

@@ -7,10 +7,9 @@ from torchaudio._internal import module_utils as _mod_utils
 from .common import AudioMetaData
 
 
-if _mod_utils.is_module_available("soundfile"):
+if _mod_utils.is_soundfile_available():
     import soundfile
 
-
 # Mapping from soundfile subtype to number of bits per sample.
 # This is mostly heuristical and the value is set to 0 when it is irrelevant
 # (lossy formats) or when it can't be inferred.
@@ -81,7 +80,7 @@ def _get_encoding(format: str, subtype: str):
     return _SUBTYPE_TO_ENCODING.get(subtype, 'UNKNOWN')
 
 
-@_mod_utils.requires_module("soundfile")
+@_mod_utils.requires_soundfile()
 def info(filepath: str, format: Optional[str] = None) -> AudioMetaData:
     """Get signal information of an audio file.
 
@@ -120,7 +119,7 @@ _SUBTYPE2DTYPE = {
 }
 
 
-@_mod_utils.requires_module("soundfile")
+@_mod_utils.requires_soundfile()
 def load(
     filepath: str,
     frame_offset: int = 0,
@@ -299,7 +298,7 @@ def _get_subtype(
     raise ValueError(f"Unsupported format: {format}")
 
 
-@_mod_utils.requires_module("soundfile")
+@_mod_utils.requires_soundfile()
 def save(
     filepath: str,
     src: torch.Tensor,