Fix for getting tokenizer in character-based ASR models when using tarred dataset

[jonghwanhyeon]

What does this PR do ?

This PR fixes getting tokenizer in character-based ASR models when using tarred dataset

Collection: ASR

Changelog

• This PR removed passing self.tokenizer to get_concat_tarred_dataset() or get_tarred_dataset()
  • Because character-based ASR models don't use tokenizers, self.tokenizer raises the following error:

Traceback (most recent call last):
  File "/workspace/asr/korean/./speech_to_text_rnnt.py", line 85, in main
    asr_model = EncDecRNNTModel(cfg=cfg.model, trainer=trainer)
  File "/workspace/NeMo/nemo/collections/asr/models/rnnt_models.py", line 53, in __init__
    super().__init__(cfg=cfg, trainer=trainer)
  File "/workspace/NeMo/nemo/core/classes/modelPT.py", line 132, in __init__
    self.setup_training_data(self._cfg.train_ds)
  File "/workspace/NeMo/nemo/collections/asr/models/rnnt_models.py", line 563, in setup_training_data
    self._train_dl = self._setup_dataloader_from_config(config=train_data_config)
  File "/workspace/NeMo/nemo/collections/asr/models/rnnt_models.py", line 509, in _setup_dataloader_from_config
    tokenizer=self.tokenizer,
  File "/usr/local/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1207, in __getattr__
    raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'EncDecRNNTModel' object has no attribute 'tokenizer'

Usage

• You can potentially add a usage example below

# Add a code snippet demonstrating how to use this 

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you add or update any necessary documentation?
• Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
  • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

• New Feature
• Bugfix
• Documentation

If you haven't finished some of the above items you can still open "Draft" PR.

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

Additional Information

• Related to # (issue)

[titu1994]

Thanks for the PR !