[TTS] Add additional config to preprocess_text and compute_feature_stats

[rlangman]

What does this PR do ?

Add some configuration to TTS preprocessing to make them more usable.

Most important, adding batch_size to text normalization multi-processing which I have found to make it about 10x faster than using the default 'auto' batch size in joblib. I do not fully understand why, but the documentation indicates that large batch sizes work for small, fast tasks:

batch_size: int or 'auto', default: 'auto'
The number of atomic tasks to dispatch at once to each
worker. When individual evaluations are very fast, dispatching
calls to workers can be slower than sequential computation because
of the overhead. Batching fast computations together can mitigate
this.

Collection: [TTS]

Changelog

• Add batch_size parameter to preprocess_text.py
• Add input and output field parameters to preprocess_text.py
• Fix boolean flag parsing for lower_case for preprocess_text.py
• Modify compute_feature_stats.py so that it can take a list of manifests, making it easy to compute pitch/energy stats across multiple datasets.

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

[racoiaws]

Otherwise LGTM