[TTS] relax hardcoded prefix for phonemes and tones and infer phoneme set through dict

[XuesongYang]

What does this PR do ?

1. previously, '#' was hardcoded as a prefix for Chinese phonemes and tones in both g2p and tokenizer. It is apt to raise errors if mismatched prefix symbols are applied for g2p and tokenizer.

2. previously, phoneme set and tone set are hardcoded as well. It would raise errors when the input phoneme dict contains different set of phonemes. Fixed this by inferring phoneme set based on input dictionary.

3. run pytest nightly and old NGC model checkpoint passed.

$ pytest --nightly tests/collections/tts/models/test_fastpitch.py::test_inference[tts_zh_fastpitch_sfspeech]

tests/collections/tts/models/test_fastpitch.py::test_inference[tts_zh_fastpitch_sfspeech] PASSED

Collection: [Note which collection this PR will affect]

Changelog

• Add specific line by line info of high level changes in this PR.

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)

[XuesongYang]

@yuekaizhang, could you please also have a look?

[yuekaizhang]

LGTM. I was wondering where you put the phonemes' list. Ignore if you would like prepare a separate PR for it.

[XuesongYang]

LGTM. I was wondering where you put the phonemes' list. Ignore if you would like prepare a separate PR for it.

The phoneme list is not predefined anymore, instead, g2p will infer one directly from the phoneme dictionary (recall that this dictionary defines the pinyin -> phonemes conversion)