Use Python 3.9 syntax in examples (#37279)

Signed-off-by: cyy <cyyever@outlook.com>

examples/pytorch/speech-recognition/run_speech_recognition_ctc.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# coding=utf-8
 # Copyright 2021 The HuggingFace Inc. team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -24,7 +23,7 @@ import re
 import sys
 import warnings
 from dataclasses import dataclass, field
-from typing import Dict, List, Optional, Union
+from typing import Optional, Union
 
 import datasets
 import evaluate
@@ -211,11 +210,11 @@ class DataTrainingArguments:
             )
         },
     )
-    chars_to_ignore: Optional[List[str]] = list_field(
+    chars_to_ignore: Optional[list[str]] = list_field(
         default=None,
         metadata={"help": "A list of characters to remove from the transcripts."},
     )
-    eval_metrics: List[str] = list_field(
+    eval_metrics: list[str] = list_field(
         default=["wer"],
         metadata={"help": "A list of metrics the model should be evaluated on. E.g. `'wer cer'`"},
     )
@@ -318,7 +317,7 @@ class DataCollatorCTCWithPadding:
     pad_to_multiple_of_labels: Optional[int] = None
     feature_extractor_input_name: Optional[str] = "input_values"
 
-    def __call__(self, features: List[Dict[str, Union[List[int], torch.Tensor]]]) -> Dict[str, torch.Tensor]:
+    def __call__(self, features: list[dict[str, Union[list[int], torch.Tensor]]]) -> dict[str, torch.Tensor]:
         # split inputs and labels since they have to be of different lengths and need
         # different padding methods
         input_features = [

examples/pytorch/speech-recognition/run_speech_recognition_ctc_adapter.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# coding=utf-8
 # Copyright 2023 The HuggingFace Inc. team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -24,7 +23,7 @@ import re
 import sys
 import warnings
 from dataclasses import dataclass, field
-from typing import Dict, List, Optional, Union
+from typing import Optional, Union
 
 import datasets
 import evaluate
@@ -201,11 +200,11 @@ class DataTrainingArguments:
             )
         },
     )
-    chars_to_ignore: Optional[List[str]] = list_field(
+    chars_to_ignore: Optional[list[str]] = list_field(
         default=None,
         metadata={"help": "A list of characters to remove from the transcripts."},
     )
-    eval_metrics: List[str] = list_field(
+    eval_metrics: list[str] = list_field(
         default=["wer"],
         metadata={"help": "A list of metrics the model should be evaluated on. E.g. `'wer cer'`"},
     )
@@ -300,7 +299,7 @@ class DataCollatorCTCWithPadding:
     pad_to_multiple_of: Optional[int] = None
     pad_to_multiple_of_labels: Optional[int] = None
 
-    def __call__(self, features: List[Dict[str, Union[List[int], torch.Tensor]]]) -> Dict[str, torch.Tensor]:
+    def __call__(self, features: list[dict[str, Union[list[int], torch.Tensor]]]) -> dict[str, torch.Tensor]:
         # split inputs and labels since they have to be of different lengths and need
         # different padding methods
         input_features = [{"input_values": feature["input_values"]} for feature in features]

examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# coding=utf-8
 # Copyright 2021 The HuggingFace Team. All rights reserved.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -23,7 +22,7 @@ import logging
 import os
 import sys
 from dataclasses import dataclass, field
-from typing import Any, Dict, List, Optional, Union
+from typing import Any, Optional, Union
 
 import datasets
 import evaluate
@@ -110,11 +109,11 @@ class ModelArguments:
     freeze_encoder: bool = field(
         default=False, metadata={"help": "Whether to freeze the entire encoder of the seq2seq model."}
     )
-    forced_decoder_ids: List[List[int]] = field(
+    forced_decoder_ids: list[list[int]] = field(
         default=None,
         metadata={"help": "Deprecated. Please use the `language` and `task` arguments instead."},
     )
-    suppress_tokens: List[int] = field(
+    suppress_tokens: list[int] = field(
         default=None,
         metadata={
             "help": (
@@ -247,7 +246,7 @@ class DataCollatorSpeechSeq2SeqWithPadding:
     decoder_start_token_id: int
     forward_attention_mask: bool
 
-    def __call__(self, features: List[Dict[str, Union[List[int], torch.Tensor]]]) -> Dict[str, torch.Tensor]:
+    def __call__(self, features: list[dict[str, Union[list[int], torch.Tensor]]]) -> dict[str, torch.Tensor]:
         # split inputs and labels since they have to be of different lengths and need
         # different padding methods
         model_input_name = self.processor.model_input_names[0]