Incorporate comments from PR huggingface#23223

src/transformers/models/audio_spectrogram_transformer/feature_extraction_audio_spectrogram_transformer.py

@@ -135,7 +135,8 @@ def __call__(
         Args:
             raw_speech (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`):
                 The sequence or batch of sequences to be padded. Each sequence can be a numpy array, a list of float
-                values, a list of numpy arrays or a list of list of float values.
+                values, a list of numpy arrays or a list of list of float values. Must be mono channel audio, not
+                stereo, i.e. single float per timestep.
             sampling_rate (`int`, *optional*):
                 The sampling rate at which the `raw_speech` input was sampled. It is strongly recommended to pass
                 `sampling_rate` at the forward call to prevent silent errors.
@@ -161,9 +162,8 @@ def __call__(
             )
 
         is_batched_numpy = isinstance(raw_speech, np.ndarray) and len(raw_speech.shape) > 1
-        if is_batched_numpy:
-            if len(raw_speech.shape) > 2:
-                raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        if is_batched_numpy and len(raw_speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
         is_batched = is_batched_numpy or (
             isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], (np.ndarray, tuple, list)))
         )

src/transformers/models/clap/feature_extraction_clap.py

@@ -272,7 +272,8 @@ def __call__(
         Args:
             raw_speech (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`):
                 The sequence or batch of sequences to be padded. Each sequence can be a numpy array, a list of float
-                values, a list of numpy arrays or a list of list of float values.
+                values, a list of numpy arrays or a list of list of float values. Must be mono channel audio, not
+                stereo, i.e. single float per timestep.
             truncation (`str`, *optional*):
                 Truncation pattern for long audio inputs. Two patterns are available:
                     - `fusion` will use `_random_mel_fusion`, which stacks 3 random crops from the mel spectrogram and
@@ -313,9 +314,8 @@ def __call__(
             )
 
         is_batched_numpy = isinstance(raw_speech, np.ndarray) and len(raw_speech.shape) > 1
-        if is_batched_numpy:
-            if len(raw_speech.shape) > 2:
-                raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        if is_batched_numpy and len(raw_speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
         is_batched = is_batched_numpy or (
             isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], (np.ndarray, tuple, list)))
         )

src/transformers/models/mctct/feature_extraction_mctct.py

@@ -180,7 +180,8 @@ def __call__(
         Args:
             raw_speech (`torch.Tensor`, `np.ndarray`, `List[float]`, `List[torch.Tensor]`, `List[np.ndarray]`, `List[List[float]]`):
                 The sequence or batch of sequences to be padded. Each sequence can be a tensor, a numpy array, a list
-                of float values, a list of tensors, a list of numpy arrays or a list of list of float values.
+                of float values, a list of tensors, a list of numpy arrays or a list of list of float values. Must be
+                mono channel audio, not stereo, i.e. single float per timestep.
             padding (`bool`, `str` or [`~file_utils.PaddingStrategy`], *optional*, defaults to `False`):
                 Select a strategy to pad the returned sequences (according to the model's padding side and padding
                 index) among:
@@ -232,9 +233,8 @@ def __call__(
             )
 
         is_batched_numpy = isinstance(raw_speech, np.ndarray) and len(raw_speech.shape) > 1
-        if is_batched_numpy:
-            if len(raw_speech.shape) > 2:
-                raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        if is_batched_numpy and len(raw_speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
         is_batched = is_batched_numpy or (
             isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], (np.ndarray, tuple, list)))
         )

src/transformers/models/speech_to_text/feature_extraction_speech_to_text.py

@@ -141,7 +141,8 @@ def __call__(
         Args:
             raw_speech (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`):
                 The sequence or batch of sequences to be padded. Each sequence can be a numpy array, a list of float
-                values, a list of numpy arrays or a list of list of float values.
+                values, a list of numpy arrays or a list of list of float values. Must be mono channel audio, not
+                stereo, i.e. single float per timestep.
             padding (`bool`, `str` or [`~utils.PaddingStrategy`], *optional*, defaults to `True`):
                 Select a strategy to pad the returned sequences (according to the model's padding side and padding
                 index) among:
@@ -201,9 +202,8 @@ def __call__(
             )
 
         is_batched_numpy = isinstance(raw_speech, np.ndarray) and len(raw_speech.shape) > 1
-        if is_batched_numpy:
-            if len(raw_speech.shape) > 2:
-                raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        if is_batched_numpy and len(raw_speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
         is_batched = is_batched_numpy or (
             isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], (np.ndarray, tuple, list)))
         )

src/transformers/models/speecht5/feature_extraction_speecht5.py

@@ -201,7 +201,8 @@ def __call__(
         Args:
             audio (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`, *optional*):
                 The sequence or batch of sequences to be processed. Each sequence can be a numpy array, a list of float
-                values, a list of numpy arrays or a list of list of float values. This outputs waveform features.
+                values, a list of numpy arrays or a list of list of float values. This outputs waveform features. Must
+                be mono channel audio, not stereo, i.e. single float per timestep.
             audio_target (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`, *optional*):
                 The sequence or batch of sequences to be processed as targets. Each sequence can be a numpy array, a
                 list of float values, a list of numpy arrays or a list of list of float values. This outputs log-mel
@@ -308,9 +309,8 @@ def _process_audio(
         **kwargs,
     ) -> BatchFeature:
         is_batched_numpy = isinstance(speech, np.ndarray) and len(speech.shape) > 1
-        if is_batched_numpy:
-            if len(speech.shape) > 2:
-                raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        if is_batched_numpy and len(speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
         is_batched = is_batched_numpy or (
             isinstance(speech, (list, tuple)) and (isinstance(speech[0], (np.ndarray, tuple, list)))
         )

src/transformers/models/tvlt/feature_extraction_tvlt.py

@@ -129,7 +129,8 @@ def __call__(
         Args:
             raw_speech (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`):
                 The sequence or batch of sequences to be padded. Each sequence can be a numpy array, a list of float
-                values, a list of numpy arrays or a list of list of float values.
+                values, a list of numpy arrays or a list of list of float values. Must be mono channel audio, not
+                stereo, i.e. single float per timestep.
             return_tensors (`str` or [`~utils.TensorType`], *optional*):
                 If set, will return tensors instead of list of python integers. Acceptable values are:
                 - `'pt'`: Return PyTorch `torch.Tensor` objects.
@@ -177,9 +178,8 @@ def __call__(
             )
 
         is_batched_numpy = isinstance(raw_speech, np.ndarray) and len(raw_speech.shape) > 1
-        if is_batched_numpy:
-            if len(raw_speech.shape) > 2:
-                raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        if is_batched_numpy and len(raw_speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
         is_batched = is_batched_numpy or (
             isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], (np.ndarray, tuple, list)))
         )

src/transformers/models/whisper/feature_extraction_whisper.py

@@ -152,7 +152,8 @@ def __call__(
         Args:
             raw_speech (`np.ndarray`, `List[float]`, `List[np.ndarray]`, `List[List[float]]`):
                 The sequence or batch of sequences to be padded. Each sequence can be a numpy array, a list of float
-                values, a list of numpy arrays or a list of list of float values.
+                values, a list of numpy arrays or a list of list of float values. Must be mono channel audio, not
+                stereo, i.e. single float per timestep.
             truncation (`bool`, *optional*, default to `True`):
                 Activates truncation to cut input sequences longer than *max_length* to *max_length*.
             pad_to_multiple_of (`int`, *optional*, defaults to None):
@@ -204,9 +205,8 @@ def __call__(
             )
 
         is_batched_numpy = isinstance(raw_speech, np.ndarray) and len(raw_speech.shape) > 1
-        if is_batched_numpy:
-            if len(raw_speech.shape) > 2:
-                raise ValueError(f"Only mono-channel audio is supported for input to {self}")
+        if is_batched_numpy and len(raw_speech.shape) > 2:
+            raise ValueError(f"Only mono-channel audio is supported for input to {self}")
         is_batched = is_batched_numpy or (
             isinstance(raw_speech, (list, tuple)) and (isinstance(raw_speech[0], (np.ndarray, tuple, list)))
         )