gptenv/neural-audio-tokenizer

AI Summary: The `neural_audio_tokenizer.py` script requires a comprehensive review and refactoring to improve code consistency and logic. A key update involves enhancing the optional Encodec sections to achieve better token diversity, drawing inspiration from `mert`'s more effective strategies beyond simple jitter. Additionally, the issue calls for modernizing command-line arguments for improved user ergonomics, with the successful completion of these tasks paving the way for a new release.