Updated the model card for wav2vec2-phoneme

[AshAnand34]

What does this PR do?

This pull request updates the documentation for the Wav2Vec2Phoneme model to improve readability, provide more examples, and clarify its usage. The changes include a revamped introduction, added usage examples, and detailed explanations of the model's features and capabilities.

Documentation Enhancements:

• Introduction and Overview: Replaced the plain text introduction with a visually appealing layout that includes badges for supported frameworks (PyTorch, TensorFlow, Flax) and a clear explanation of the Wav2Vec2Phoneme model's purpose and capabilities. Added links to relevant papers and model checkpoints.

• Usage Examples: Introduced two detailed code examples for phoneme recognition using the Pipeline and AutoModel classes. These examples demonstrate how to load the model, process audio data, and decode phonemes.

• Quantization Support: Added a section on quantization, including an example of using bitsandbytes to load the model with 8-bit precision, reducing memory usage.

Detailed Feature Descriptions:

• Model Features: Expanded on the model's architecture, cross-lingual capabilities, and performance. Explained its ability to map phonemes across languages and its use of connectionist temporal classification (CTC) for training. (Ff214

#36979

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@stevhliu