diff --git a/src/transformers/models/hubert/modeling_hubert.py b/src/transformers/models/hubert/modeling_hubert.py
index f0f31e87716e..59b199e26d27 100755
--- a/src/transformers/models/hubert/modeling_hubert.py
+++ b/src/transformers/models/hubert/modeling_hubert.py
@@ -41,14 +41,10 @@
 
 logger = logging.get_logger(__name__)
 
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
-
-
 _HIDDEN_STATES_START_POSITION = 1
 
 # General docstring
 _CONFIG_FOR_DOC = "HubertConfig"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "facebook/hubert-large-ls960-ft"
@@ -59,7 +55,6 @@
 _CTC_EXPECTED_LOSS = 22.68
 
 # Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
 _SEQ_CLASS_CHECKPOINT = "superb/hubert-base-superb-ks"
 _SEQ_CLASS_EXPECTED_OUTPUT = "'_unknown_'"
 _SEQ_CLASS_EXPECTED_LOSS = 8.53
@@ -1145,7 +1140,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(HUBERT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1280,7 +1274,6 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(HUBERT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_SEQ_CLASS_CHECKPOINT,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
diff --git a/src/transformers/models/longformer/modeling_longformer.py b/src/transformers/models/longformer/modeling_longformer.py
index 137e99a67c01..606459067df5 100755
--- a/src/transformers/models/longformer/modeling_longformer.py
+++ b/src/transformers/models/longformer/modeling_longformer.py
@@ -41,7 +41,6 @@
 
 _CHECKPOINT_FOR_DOC = "allenai/longformer-base-4096"
 _CONFIG_FOR_DOC = "LongformerConfig"
-_TOKENIZER_FOR_DOC = "LongformerTokenizer"
 
 LONGFORMER_PRETRAINED_MODEL_ARCHIVE_LIST = [
     "allenai/longformer-base-4096",
@@ -1903,7 +1902,6 @@ def __init__(self, config):
 
     @add_start_docstrings_to_model_forward(LONGFORMER_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="jpwahle/longformer-base-plagiarism-detection",
         output_type=LongformerSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2172,7 +2170,6 @@ def __init__(self, config):
 
     @add_start_docstrings_to_model_forward(LONGFORMER_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint="brad1141/Longformer-finetuned-norm",
         output_type=LongformerTokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2260,7 +2257,6 @@ def __init__(self, config):
         LONGFORMER_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length")
     )
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=LongformerMultipleChoiceModelOutput,
         config_class=_CONFIG_FOR_DOC,
diff --git a/src/transformers/models/longformer/modeling_tf_longformer.py b/src/transformers/models/longformer/modeling_tf_longformer.py
index 7c929f353d30..b50231f659f9 100644
--- a/src/transformers/models/longformer/modeling_tf_longformer.py
+++ b/src/transformers/models/longformer/modeling_tf_longformer.py
@@ -2382,11 +2382,9 @@ def __init__(self, config, *inputs, **kwargs):
     @add_start_docstrings_to_model_forward(LONGFORMER_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
         processor_class=_TOKENIZER_FOR_DOC,
-        checkpoint="hf-internal-testing/tiny-random-longformer",
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFLongformerSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
-        expected_output="'LABEL_1'",
-        expected_loss=0.69,
     )
     def call(
         self,
@@ -2637,15 +2635,9 @@ def __init__(self, config, *inputs, **kwargs):
     @add_start_docstrings_to_model_forward(LONGFORMER_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
     @add_code_sample_docstrings(
         processor_class=_TOKENIZER_FOR_DOC,
-        checkpoint="hf-internal-testing/tiny-random-longformer",
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TFLongformerTokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
-        expected_output=(
-            "['LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1',"
-            " 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1', 'LABEL_1',"
-            " 'LABEL_1', 'LABEL_1']"
-        ),
-        expected_loss=0.59,
     )
     def call(
         self,
diff --git a/src/transformers/models/mbart/modeling_mbart.py b/src/transformers/models/mbart/modeling_mbart.py
index 421393195e44..1cf3b00ce328 100755
--- a/src/transformers/models/mbart/modeling_mbart.py
+++ b/src/transformers/models/mbart/modeling_mbart.py
@@ -49,22 +49,10 @@
 
 _CHECKPOINT_FOR_DOC = "facebook/mbart-large-cc25"
 _CONFIG_FOR_DOC = "MBartConfig"
-_TOKENIZER_FOR_DOC = "MBartTokenizer"
 
 # Base model docstring
 _EXPECTED_OUTPUT_SHAPE = [1, 8, 1024]
 
-# SequenceClassification docstring
-_CHECKPOINT_FOR_SEQUENCE_CLASSIFICATION = "hf-internal-testing/tiny-random-mbart"
-_SEQ_CLASS_EXPECTED_LOSS = 0.69
-_SEQ_CLASS_EXPECTED_OUTPUT = "'LABEL_1'"
-
-# QuestionAsnwering docstring
-_CHECKPOINT_FOR_QA = "hf-internal-testing/tiny-random-mbart"
-_QA_EXPECTED_LOSS = 3.55
-_QA_EXPECTED_OUTPUT = "'? Jim Henson was a'"
-
-
 MBART_PRETRAINED_MODEL_ARCHIVE_LIST = [
     "facebook/mbart-large-cc25",
     # See all MBART models at https://huggingface.co/models?filter=mbart
@@ -1187,7 +1175,6 @@ def get_decoder(self):
 
     @add_start_docstrings_to_model_forward(MBART_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1467,12 +1454,9 @@ def __init__(self, config: MBartConfig, **kwargs):
 
     @add_start_docstrings_to_model_forward(MBART_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
-        checkpoint=_CHECKPOINT_FOR_SEQUENCE_CLASSIFICATION,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
-        expected_output=_SEQ_CLASS_EXPECTED_OUTPUT,
-        expected_loss=_SEQ_CLASS_EXPECTED_LOSS,
     )
     # Copied from transformers.models.bart.modeling_bart.BartForSequenceClassification.forward
     def forward(
@@ -1596,12 +1580,9 @@ def __init__(self, config):
 
     @add_start_docstrings_to_model_forward(MBART_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
-        checkpoint=_CHECKPOINT_FOR_QA,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqQuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,
-        expected_loss=_QA_EXPECTED_LOSS,
-        expected_output=_QA_EXPECTED_OUTPUT,
     )
     # Copied from transformers.models.bart.modeling_bart.BartForQuestionAnswering.forward
     def forward(
diff --git a/src/transformers/models/plbart/modeling_plbart.py b/src/transformers/models/plbart/modeling_plbart.py
index 1e2bc3bcc655..98017c48c31b 100644
--- a/src/transformers/models/plbart/modeling_plbart.py
+++ b/src/transformers/models/plbart/modeling_plbart.py
@@ -48,16 +48,6 @@
 
 _CHECKPOINT_FOR_DOC = "uclanlp/plbart-base"
 _CONFIG_FOR_DOC = "PLBartConfig"
-_TOKENIZER_FOR_DOC = "PLBartTokenizer"
-
-# Base model docstring
-_EXPECTED_OUTPUT_SHAPE = [1, 8, 768]
-
-# SequenceClassification docstring
-_CHECKPOINT_FOR_SEQUENCE_CLASSIFICATION = "hf-internal-testing/tiny-plbart"
-_SEQ_CLASS_EXPECTED_OUTPUT = "'LABEL_1'"
-_SEQ_CLASS_EXPECTED_LOSS = 0.69
-
 
 PLBART_PRETRAINED_MODEL_ARCHIVE_LIST = [
     "uclanlp/plbart-base",
@@ -1161,7 +1151,6 @@ def get_decoder(self):
 
     @add_start_docstrings_to_model_forward(PLBART_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1438,12 +1427,9 @@ def __init__(self, config: PLBartConfig, **kwargs):
 
     @add_start_docstrings_to_model_forward(PLBART_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
-        checkpoint=_CHECKPOINT_FOR_SEQUENCE_CLASSIFICATION,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Seq2SeqSequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
-        expected_output=_SEQ_CLASS_EXPECTED_OUTPUT,
-        expected_loss=_SEQ_CLASS_EXPECTED_LOSS,
     )
     # Copied from transformers.models.bart.modeling_bart.BartForSequenceClassification.forward
     def forward(
diff --git a/src/transformers/models/reformer/modeling_reformer.py b/src/transformers/models/reformer/modeling_reformer.py
index 0cca45422eaf..3ceaf24a42e2 100755
--- a/src/transformers/models/reformer/modeling_reformer.py
+++ b/src/transformers/models/reformer/modeling_reformer.py
@@ -49,7 +49,6 @@
 
 _CHECKPOINT_FOR_DOC = "google/reformer-crime-and-punishment"
 _CONFIG_FOR_DOC = "ReformerConfig"
-_TOKENIZER_FOR_DOC = "ReformerTokenizer"
 
 REFORMER_PRETRAINED_MODEL_ARCHIVE_LIST = [
     "google/reformer-crime-and-punishment",
@@ -2009,7 +2008,6 @@ class PreTrainedModel
 
     @add_start_docstrings_to_model_forward(REFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=ReformerModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2220,7 +2218,6 @@ def set_output_embeddings(self, new_embeddings):
 
     @add_start_docstrings_to_model_forward(REFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2360,13 +2357,20 @@ def forward(
 
         Returns:
 
+        <Tip warning={true}>
+
+        This example uses a false checkpoint since we don't have any available pretrained model for the masked language
+        modeling task with the Reformer architecture.
+
+        </Tip>
+
         Example:
 
         ```python
         >>> import torch
-        >>> from transformers import ReformerTokenizer, ReformerForMaskedLM
+        >>> from transformers import AutoTokenizer, ReformerForMaskedLM
 
-        >>> tokenizer = ReformerTokenizer.from_pretrained("hf-internal-testing/tiny-random-reformer")
+        >>> tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/tiny-random-reformer")
         >>> model = ReformerForMaskedLM.from_pretrained("hf-internal-testing/tiny-random-reformer")
 
         >>> # add mask_token
@@ -2479,10 +2483,10 @@ def forward(
 
         ```python
         >>> import torch
-        >>> from transformers import ReformerTokenizer, ReformerForSequenceClassification
+        >>> from transformers import AutoTokenizer, ReformerForSequenceClassification
 
-        >>> tokenizer = ReformerTokenizer.from_pretrained("hf-internal-testing/tiny-random-reformer")
-        >>> model = ReformerForSequenceClassification.from_pretrained("hf-internal-testing/tiny-random-reformer")
+        >>> tokenizer = AutoTokenizer.from_pretrained("google/reformer-crime-and-punishment")
+        >>> model = ReformerForSequenceClassification.from_pretrained("google/reformer-crime-and-punishment")
 
         >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
 
@@ -2491,59 +2495,20 @@ def forward(
 
         >>> predicted_class_id = logits.argmax().item()
         >>> model.config.id2label[predicted_class_id]
-        'LABEL_1'
+        'LABEL_0'
         ```
 
         ```python
         >>> # To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)`
         >>> num_labels = len(model.config.id2label)
         >>> model = ReformerForSequenceClassification.from_pretrained(
-        ...     "hf-internal-testing/tiny-random-reformer", num_labels=num_labels
+        ...     "google/reformer-crime-and-punishment", num_labels=num_labels
         ... )
 
         >>> labels = torch.tensor(1)
         >>> loss = model(**inputs, labels=labels).loss
         >>> round(loss.item(), 2)
-        0.69
-        ```
-
-        Example of multi-label classification:
-
-        ```python
-        >>> import torch
-        >>> from transformers import ReformerTokenizer, ReformerForSequenceClassification
-
-        >>> tokenizer = ReformerTokenizer.from_pretrained("hf-internal-testing/tiny-random-reformer")
-        >>> model = ReformerForSequenceClassification.from_pretrained(
-        ...     "hf-internal-testing/tiny-random-reformer", problem_type="multi_label_classification"
-        ... )
-
-        >>> # add pad_token
-        >>> tokenizer.add_special_tokens({"pad_token": "[PAD]"})  # doctest: +IGNORE_RESULT
-        >>> inputs = tokenizer("Hello, my dog is cute", max_length=100, padding="max_length", return_tensors="pt")
-
-        >>> with torch.no_grad():
-        ...     logits = model(**inputs).logits
-
-        >>> predicted_class_id = logits.argmax().item()
-        >>> model.config.id2label[predicted_class_id]
-        'LABEL_1'
-        ```
-
-        ```python
-        >>> # To train a model on `num_labels` classes, you can pass `num_labels=num_labels` to `.from_pretrained(...)`
-        >>> num_labels = len(model.config.id2label)
-        >>> model = ReformerForSequenceClassification.from_pretrained(
-        ...     "hf-internal-testing/tiny-random-reformer", num_labels=num_labels
-        ... )
-        >>> model.train()  # doctest: +IGNORE_RESULT
-
-        >>> num_labels = len(model.config.id2label)
-        >>> labels = torch.nn.functional.one_hot(torch.tensor([predicted_class_id]), num_classes=num_labels).to(
-        ...     torch.float
-        ... )
-        >>> loss = model(**inputs, labels=labels).loss
-        >>> loss.backward()  # doctest: +IGNORE_RESULT
+        0.68
         ```
         """
         return_dict = return_dict if return_dict is not None else self.config.use_return_dict
@@ -2641,12 +2606,9 @@ def __init__(self, config):
 
     @add_start_docstrings_to_model_forward(REFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_TOKENIZER_FOR_DOC,
-        checkpoint="hf-internal-testing/tiny-random-reformer",
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=QuestionAnsweringModelOutput,
         config_class=_CONFIG_FOR_DOC,
-        expected_output="''",
-        expected_loss=3.28,
     )
     def forward(
         self,
diff --git a/src/transformers/models/sew/modeling_sew.py b/src/transformers/models/sew/modeling_sew.py
index bed9e9fbdf0c..e358c75bb557 100644
--- a/src/transformers/models/sew/modeling_sew.py
+++ b/src/transformers/models/sew/modeling_sew.py
@@ -36,16 +36,11 @@
 
 logger = logging.get_logger(__name__)
 
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
-
 
 _HIDDEN_STATES_START_POSITION = 1
 
-
 # General docstring
 _CONFIG_FOR_DOC = "SEWConfig"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "asapp/sew-tiny-100k-ft-ls100h"
@@ -58,7 +53,6 @@
 _CTC_EXPECTED_LOSS = 0.42
 
 # Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
 _SEQ_CLASS_CHECKPOINT = "anton-l/sew-mid-100k-ft-keyword-spotting"
 _SEQ_CLASS_EXPECTED_OUTPUT = "'_unknown_'"
 _SEQ_CLASS_EXPECTED_LOSS = 9.52
@@ -916,7 +910,6 @@ def _mask_hidden_states(
 
     @add_start_docstrings_to_model_forward(SEW_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1020,7 +1013,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(SEW_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1155,7 +1147,6 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(SEW_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_SEQ_CLASS_CHECKPOINT,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
diff --git a/src/transformers/models/sew_d/modeling_sew_d.py b/src/transformers/models/sew_d/modeling_sew_d.py
index 804f78c26cdf..64c7b006363d 100644
--- a/src/transformers/models/sew_d/modeling_sew_d.py
+++ b/src/transformers/models/sew_d/modeling_sew_d.py
@@ -42,7 +42,6 @@
 
 # General docstring
 _CONFIG_FOR_DOC = "SEWDConfig"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "asapp/sew-d-tiny-100k-ft-ls100h"
@@ -53,7 +52,6 @@
 _CTC_EXPECTED_LOSS = 0.21
 
 # Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
 _SEQ_CLASS_CHECKPOINT = "anton-l/sew-d-mid-400k-ft-keyword-spotting"
 _SEQ_CLASS_EXPECTED_OUTPUT = "'_unknown_'"
 _SEQ_CLASS_EXPECTED_LOSS = 3.16
@@ -1453,7 +1451,6 @@ def _mask_hidden_states(
 
     @add_start_docstrings_to_model_forward(SEWD_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1557,7 +1554,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(SEWD_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1692,7 +1688,6 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(SEWD_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_SEQ_CLASS_CHECKPOINT,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
diff --git a/src/transformers/models/unispeech/modeling_unispeech.py b/src/transformers/models/unispeech/modeling_unispeech.py
index 2949a0201963..11d7a30ad9c8 100755
--- a/src/transformers/models/unispeech/modeling_unispeech.py
+++ b/src/transformers/models/unispeech/modeling_unispeech.py
@@ -48,7 +48,6 @@
 
 # General docstring
 _CONFIG_FOR_DOC = "UniSpeechConfig"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "patrickvonplaten/unispeech-large-1500h-cv-timit"
@@ -58,12 +57,6 @@
 _CTC_EXPECTED_OUTPUT = "'mister quilter is the apposl of the midle classes and weare glad to welcom his gosepl'"
 _CTC_EXPECTED_LOSS = 17.17
 
-# Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
-_SEQ_CLASS_CHECKPOINT = "hf-internal-testing/tiny-random-unispeech"
-_SEQ_CLASS_EXPECTED_OUTPUT = "'LABEL_0'"  # TODO(anton) - could you quickly fine-tune a KS WavLM Model
-_SEQ_CLASS_EXPECTED_LOSS = 0.66  # TODO(anton) - could you quickly fine-tune a KS WavLM Model
-
 UNISPEECH_PRETRAINED_MODEL_ARCHIVE_LIST = [
     "microsoft/unispeech-large-1500h-cv",
     "microsoft/unispeech-large-multi-lingual-1500h-cv",
@@ -1143,7 +1136,6 @@ def _mask_hidden_states(
 
     @add_start_docstrings_to_model_forward(UNISPEECH_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Wav2Vec2BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1286,12 +1278,9 @@ def forward(
 
         ```python
         >>> import torch
-        >>> from transformers import Wav2Vec2FeatureExtractor, UniSpeechForPreTraining
-        >>> from transformers.models.unispeech.modeling_unispeech import _compute_mask_indices
+        >>> from transformers import AutoFeatureExtractor, UniSpeechForPreTraining
 
-        >>> feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(
-        ...     "hf-internal-testing/tiny-random-unispeech-sat"
-        ... )
+        >>> feature_extractor = AutoFeatureExtractor.from_pretrained("microsoft/unispeech-large-1500h-cv")
         >>> model = UniSpeechForPreTraining.from_pretrained("microsoft/unispeech-large-1500h-cv")
         >>> # TODO: Add full pretraining example
         ```"""
@@ -1395,7 +1384,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(UNISPEECH_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1482,7 +1470,6 @@ def forward(
     """,
     UNISPEECH_START_DOCSTRING,
 )
-# Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification with Wav2Vec2->UniSpeech, wav2vec2->unispeech, WAV_2_VEC_2->UNISPEECH
 class UniSpeechForSequenceClassification(UniSpeechPreTrainedModel):
     def __init__(self, config):
         super().__init__(config)
@@ -1501,6 +1488,7 @@ def __init__(self, config):
         # Initialize weights and apply final processing
         self.post_init()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_feature_extractor
     def freeze_feature_extractor(self):
         """
         Calling this function will disable the gradient computation for the feature encoder so that its parameters will
@@ -1513,6 +1501,7 @@ def freeze_feature_extractor(self):
         )
         self.freeze_feature_encoder()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_feature_encoder with wav2vec2->unispeech
     def freeze_feature_encoder(self):
         """
         Calling this function will disable the gradient computation for the feature encoder so that its parameter will
@@ -1520,6 +1509,7 @@ def freeze_feature_encoder(self):
         """
         self.unispeech.feature_extractor._freeze_parameters()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_base_model with wav2vec2->unispeech
     def freeze_base_model(self):
         """
         Calling this function will disable the gradient computation for the base model so that its parameters will not
@@ -1530,14 +1520,12 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(UNISPEECH_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
-        checkpoint=_SEQ_CLASS_CHECKPOINT,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
         modality="audio",
-        expected_output=_SEQ_CLASS_EXPECTED_OUTPUT,
-        expected_loss=_SEQ_CLASS_EXPECTED_LOSS,
     )
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.forward with Wav2Vec2->UniSpeech, wav2vec2->unispeech
     def forward(
         self,
         input_values: Optional[torch.Tensor],
diff --git a/src/transformers/models/unispeech_sat/modeling_unispeech_sat.py b/src/transformers/models/unispeech_sat/modeling_unispeech_sat.py
index d1b2074531ed..045da6231b5b 100755
--- a/src/transformers/models/unispeech_sat/modeling_unispeech_sat.py
+++ b/src/transformers/models/unispeech_sat/modeling_unispeech_sat.py
@@ -55,7 +55,6 @@
 
 # General docstring
 _CONFIG_FOR_DOC = "UniSpeechSatConfig"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "microsoft/unispeech-sat-base-100h-libri-ft"
@@ -65,12 +64,6 @@
 _CTC_EXPECTED_OUTPUT = "'MISTER QUILDER IS THE APOSTLE OF THE MIDDLE CLASSES AND WE ARE GLAD TO WELCOME HIS GOSPEL'"
 _CTC_EXPECTED_LOSS = 39.88
 
-# Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
-_SEQ_CLASS_CHECKPOINT = "hf-internal-testing/tiny-random-unispeech-sat"
-_SEQ_CLASS_EXPECTED_OUTPUT = "'LABEL_1'"  # TODO(anton) - could you quickly fine-tune a KS WavLM Model
-_SEQ_CLASS_EXPECTED_LOSS = 0.71  # TODO(anton) - could you quickly fine-tune a KS WavLM Model
-
 # Frame class docstring
 _FRAME_CLASS_CHECKPOINT = "microsoft/unispeech-sat-base-plus-sd"
 _FRAME_EXPECTED_OUTPUT = [0, 0]
@@ -1158,7 +1151,6 @@ def _mask_hidden_states(
 
     @add_start_docstrings_to_model_forward(UNISPEECH_SAT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Wav2Vec2BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1299,10 +1291,10 @@ def forward(
 
         ```python
         >>> import torch
-        >>> from transformers import Wav2Vec2FeatureExtractor, UniSpeechSatForPreTraining
+        >>> from transformers import AutoFeatureExtractor, UniSpeechSatForPreTraining
         >>> from transformers.models.unispeech_sat.modeling_unispeech_sat import _compute_mask_indices
 
-        >>> feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained("microsoft/unispeech-sat-base")
+        >>> feature_extractor = AutoFeatureExtractor.from_pretrained("microsoft/unispeech-sat-base")
         >>> model = UniSpeechSatForPreTraining.from_pretrained("microsoft/unispeech-sat-base")
         >>> # TODO: Add full pretraining example
         ```"""
@@ -1399,7 +1391,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(UNISPEECH_SAT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1486,7 +1477,6 @@ def forward(
     """,
     UNISPEECH_SAT_START_DOCSTRING,
 )
-# Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification with Wav2Vec2->UniSpeechSat, wav2vec2->unispeech_sat, WAV_2_VEC_2->UNISPEECH_SAT
 class UniSpeechSatForSequenceClassification(UniSpeechSatPreTrainedModel):
     def __init__(self, config):
         super().__init__(config)
@@ -1505,6 +1495,7 @@ def __init__(self, config):
         # Initialize weights and apply final processing
         self.post_init()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_feature_extractor
     def freeze_feature_extractor(self):
         """
         Calling this function will disable the gradient computation for the feature encoder so that its parameters will
@@ -1517,6 +1508,7 @@ def freeze_feature_extractor(self):
         )
         self.freeze_feature_encoder()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_feature_encoder with wav2vec2->unispeech_sat
     def freeze_feature_encoder(self):
         """
         Calling this function will disable the gradient computation for the feature encoder so that its parameter will
@@ -1524,6 +1516,7 @@ def freeze_feature_encoder(self):
         """
         self.unispeech_sat.feature_extractor._freeze_parameters()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_base_model with wav2vec2->unispeech_sat
     def freeze_base_model(self):
         """
         Calling this function will disable the gradient computation for the base model so that its parameters will not
@@ -1534,14 +1527,12 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(UNISPEECH_SAT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
-        checkpoint=_SEQ_CLASS_CHECKPOINT,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
         modality="audio",
-        expected_output=_SEQ_CLASS_EXPECTED_OUTPUT,
-        expected_loss=_SEQ_CLASS_EXPECTED_LOSS,
     )
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.forward with Wav2Vec2->UniSpeechSat, wav2vec2->unispeech_sat
     def forward(
         self,
         input_values: Optional[torch.Tensor],
@@ -1658,7 +1649,6 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(UNISPEECH_SAT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_FRAME_CLASS_CHECKPOINT,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1841,7 +1831,6 @@ def _conv_out_length(input_length, kernel_size, stride):
 
     @add_start_docstrings_to_model_forward(UNISPEECH_SAT_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_XVECTOR_CHECKPOINT,
         output_type=XVectorOutput,
         config_class=_CONFIG_FOR_DOC,
diff --git a/src/transformers/models/wav2vec2/modeling_wav2vec2.py b/src/transformers/models/wav2vec2/modeling_wav2vec2.py
index cb2aeb7562ef..2a88147e3219 100755
--- a/src/transformers/models/wav2vec2/modeling_wav2vec2.py
+++ b/src/transformers/models/wav2vec2/modeling_wav2vec2.py
@@ -56,7 +56,6 @@
 
 # General docstring
 _CONFIG_FOR_DOC = "Wav2Vec2Config"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "facebook/wav2vec2-base-960h"
@@ -67,7 +66,6 @@
 _CTC_EXPECTED_LOSS = 53.48
 
 # Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
 _SEQ_CLASS_CHECKPOINT = "superb/wav2vec2-base-superb-ks"
 _SEQ_CLASS_EXPECTED_OUTPUT = "'_unknown_'"
 _SEQ_CLASS_EXPECTED_LOSS = 6.54
@@ -1279,7 +1277,6 @@ def _mask_hidden_states(
 
     @add_start_docstrings_to_model_forward(WAV_2_VEC_2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Wav2Vec2BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1655,7 +1652,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(WAV_2_VEC_2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1789,7 +1785,6 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(WAV_2_VEC_2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_SEQ_CLASS_CHECKPOINT,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1911,7 +1906,6 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(WAV_2_VEC_2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_FRAME_CLASS_CHECKPOINT,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -2091,7 +2085,6 @@ def _conv_out_length(input_length, kernel_size, stride):
 
     @add_start_docstrings_to_model_forward(WAV_2_VEC_2_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_XVECTOR_CHECKPOINT,
         output_type=XVectorOutput,
         config_class=_CONFIG_FOR_DOC,
diff --git a/src/transformers/models/wav2vec2_conformer/modeling_wav2vec2_conformer.py b/src/transformers/models/wav2vec2_conformer/modeling_wav2vec2_conformer.py
index d72522d294be..5649f4c81ed3 100644
--- a/src/transformers/models/wav2vec2_conformer/modeling_wav2vec2_conformer.py
+++ b/src/transformers/models/wav2vec2_conformer/modeling_wav2vec2_conformer.py
@@ -54,7 +54,6 @@
 
 # General docstring
 _CONFIG_FOR_DOC = "Wav2Vec2ConformerConfig"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "facebook/wav2vec2-conformer-rope-large-960h-ft"
@@ -64,20 +63,6 @@
 _CTC_EXPECTED_OUTPUT = "'MISTER QUILTER IS THE APOSTLE OF THE MIDDLE CLASSES AND WE ARE GLAD TO WELCOME HIS GOSPEL'"
 _CTC_EXPECTED_LOSS = 64.21
 
-# Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
-_SEQ_CLASS_CHECKPOINT = "hf-internal-testing/wav2vec2-conformer-seq-class"
-_SEQ_CLASS_EXPECTED_OUTPUT = "'LABEL_0'"
-_SEQ_CLASS_EXPECTED_LOSS = 0.68
-
-# Frame class docstring
-_FRAME_CLASS_CHECKPOINT = "hf-internal-testing/wav2vec2-conformer-frame-class"
-_FRAME_EXPECTED_OUTPUT = [1, 0]
-
-# Speaker Verification docstring
-_XVECTOR_CHECKPOINT = "hf-internal-testing/wav2vec2-conformer-xvector"
-_XVECTOR_EXPECTED_OUTPUT = 1.0
-
 
 WAV2VEC2_CONFORMER_PRETRAINED_MODEL_ARCHIVE_LIST = [
     "facebook/wav2vec2-conformer-rel-pos-large",
@@ -1324,7 +1309,6 @@ def _mask_hidden_states(
 
     @add_start_docstrings_to_model_forward(WAV2VEC2_CONFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Wav2Vec2BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1643,7 +1627,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(WAV2VEC2_CONFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1769,13 +1752,10 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(WAV2VEC2_CONFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
-        checkpoint=_SEQ_CLASS_CHECKPOINT,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
         modality="audio",
-        expected_output=_SEQ_CLASS_EXPECTED_OUTPUT,
-        expected_loss=_SEQ_CLASS_EXPECTED_LOSS,
     )
     # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.forward with Wav2Vec2->Wav2Vec2Conformer,wav2vec2->wav2vec2_conformer,WAV_2_VEC_2->WAV2VEC2_CONFORMER
     def forward(
@@ -1884,12 +1864,10 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(WAV2VEC2_CONFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
-        checkpoint=_FRAME_CLASS_CHECKPOINT,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
         modality="audio",
-        expected_output=_FRAME_EXPECTED_OUTPUT,
     )
     # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForAudioFrameClassification.forward with wav2vec2->wav2vec2_conformer
     def forward(
@@ -2058,12 +2036,10 @@ def _conv_out_length(input_length, kernel_size, stride):
 
     @add_start_docstrings_to_model_forward(WAV2VEC2_CONFORMER_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
-        checkpoint=_XVECTOR_CHECKPOINT,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=XVectorOutput,
         config_class=_CONFIG_FOR_DOC,
         modality="audio",
-        expected_output=_XVECTOR_EXPECTED_OUTPUT,
     )
     # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForXVector.forward with Wav2Vec2->Wav2Vec2Conformer,wav2vec2->wav2vec2_conformer,WAV_2_VEC_2->WAV2VEC2_CONFORMER
     def forward(
diff --git a/src/transformers/models/wavlm/modeling_wavlm.py b/src/transformers/models/wavlm/modeling_wavlm.py
index 2d77c6a33e37..30cec09efac5 100755
--- a/src/transformers/models/wavlm/modeling_wavlm.py
+++ b/src/transformers/models/wavlm/modeling_wavlm.py
@@ -48,7 +48,6 @@
 
 # General docstring
 _CONFIG_FOR_DOC = "WavLMConfig"
-_PROCESSOR_FOR_DOC = "Wav2Vec2Processor"
 
 # Base docstring
 _CHECKPOINT_FOR_DOC = "patrickvonplaten/wavlm-libri-clean-100h-base-plus"
@@ -58,12 +57,6 @@
 _CTC_EXPECTED_OUTPUT = "'mister quilter is the aposle of the middle classes and we are glad to welcome his gospel'"
 _CTC_EXPECTED_LOSS = 12.51
 
-# Audio class docstring
-_FEAT_EXTRACTOR_FOR_DOC = "Wav2Vec2FeatureExtractor"
-_SEQ_CLASS_CHECKPOINT = "hf-internal-testing/tiny-random-wavlm"
-_SEQ_CLASS_EXPECTED_OUTPUT = "'no'"  # TODO(anton) - could you quickly fine-tune a KS WavLM Model
-_SEQ_CLASS_EXPECTED_LOSS = 0.7  # TODO(anton) - could you quickly fine-tune a KS WavLM Model
-
 # Frame class docstring
 _FRAME_CLASS_CHECKPOINT = "microsoft/wavlm-base-plus-sd"
 _FRAME_EXPECTED_OUTPUT = [0, 0]
@@ -1212,7 +1205,6 @@ def _mask_hidden_states(
 
     @add_start_docstrings_to_model_forward(WAVLM_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=Wav2Vec2BaseModelOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1320,7 +1312,6 @@ def freeze_feature_encoder(self):
 
     @add_start_docstrings_to_model_forward(WAVLM_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_PROCESSOR_FOR_DOC,
         checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=CausalLMOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1407,7 +1398,6 @@ def forward(
     """,
     WAVLM_START_DOCSTRING,
 )
-# Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification with Wav2Vec2->WavLM, wav2vec2->wavlm, WAV_2_VEC_2->WAVLM
 class WavLMForSequenceClassification(WavLMPreTrainedModel):
     def __init__(self, config):
         super().__init__(config)
@@ -1426,6 +1416,7 @@ def __init__(self, config):
         # Initialize weights and apply final processing
         self.post_init()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_feature_extractor
     def freeze_feature_extractor(self):
         """
         Calling this function will disable the gradient computation for the feature encoder so that its parameters will
@@ -1438,6 +1429,7 @@ def freeze_feature_extractor(self):
         )
         self.freeze_feature_encoder()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_feature_encoder with wav2vec2->wavlm
     def freeze_feature_encoder(self):
         """
         Calling this function will disable the gradient computation for the feature encoder so that its parameter will
@@ -1445,6 +1437,7 @@ def freeze_feature_encoder(self):
         """
         self.wavlm.feature_extractor._freeze_parameters()
 
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.freeze_base_model with wav2vec2->wavlm
     def freeze_base_model(self):
         """
         Calling this function will disable the gradient computation for the base model so that its parameters will not
@@ -1455,14 +1448,12 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(WAVLM_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
-        checkpoint=_SEQ_CLASS_CHECKPOINT,
+        checkpoint=_CHECKPOINT_FOR_DOC,
         output_type=SequenceClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
         modality="audio",
-        expected_output=_SEQ_CLASS_EXPECTED_OUTPUT,
-        expected_loss=_SEQ_CLASS_EXPECTED_LOSS,
     )
+    # Copied from transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForSequenceClassification.forward with Wav2Vec2->WavLM, wav2vec2->wavlm
     def forward(
         self,
         input_values: Optional[torch.Tensor],
@@ -1578,7 +1569,6 @@ def freeze_base_model(self):
 
     @add_start_docstrings_to_model_forward(WAVLM_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_FRAME_CLASS_CHECKPOINT,
         output_type=TokenClassifierOutput,
         config_class=_CONFIG_FOR_DOC,
@@ -1761,7 +1751,6 @@ def _conv_out_length(input_length, kernel_size, stride):
 
     @add_start_docstrings_to_model_forward(WAVLM_INPUTS_DOCSTRING)
     @add_code_sample_docstrings(
-        processor_class=_FEAT_EXTRACTOR_FOR_DOC,
         checkpoint=_XVECTOR_CHECKPOINT,
         output_type=XVectorOutput,
         config_class=_CONFIG_FOR_DOC,