[Bug] PhonemeCoverage is out of date with the current APIs

[jhwgh1968]

Describe the bug

The notebook does not work due to a number of imports that don't exist.

I opened a trivial PR for this here: #2840 but I decline to sign the CLA so I closed it.

Perhaps someone who can will make this trivial fix and save annoyance by users trying to debug custom models?

To Reproduce

1. Open the notebook for Phoneme coverage
2. Change the config path to one being debugged
3. Run it

Expected behavior

It runs correctly and outputs an image of a phoneme distribution

Logs

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
Cell In[28], line 2
      1 # import stuff
----> 2 from TTS.utils.io import load_config
      3 from TTS.tts.datasets.formatters import load_tts_samples
      4 from TTS.tts.utils.text import phoneme_to_sequence, sequence_to_phoneme

ImportError: cannot import name 'load_config' from 'TTS.utils.io' (/home/cryptkeeper/Desktop/Software/TTS/TTS/utils/io.py)

Environment

- TTS version: git dev 4e7f8cd0
- Python version: 3.11.3

Additional context

No response