Create lexicon and HMM model for post-voice-model language processing

[Menjadianjay]

Create a lexicon and Hidden Markov Model (HMM) module for processing words and sentences extracted from a voice model. The goal is to build reusable components that allow further natural language processing and recognition tasks on the output of an existing voice-to-text system.

Tasks:

• Design and implement a lexicon to map recognized phonemes to words
• Build an HMM model for context-aware sequence modeling of extracted words/sentences
• Integrate the lexicon and HMM so they work together to process the voice model's output
• Write documentation and usage examples for working with words and sentences after voice extraction

Acceptance Criteria:

• Able to process a sequence of tokens/words from a voice model and output normalized sentences
• Lexicon and HMM components are modular and well-documented
• Example scripts for inference and demonstration are provided

Scope Constraints:

• This issue does not include voice model development; expects already extracted words/sentences as input