Refactor core queue initialization: move queue definitions to `constants.py` and clean up redundant imports in `main.py` for better modularity.

Introduce initial API structure for VoiceDialogue: add dependencies, middleware, and routes for ASR, TTS, system, and voice modules.

Update dependencies: add FunASR, FunASR-ONNX, FastAPI, and Uvicorn to requirements.txt

Refactor ASR module: introduce modular structure with ASR interface, implement FunASR and Whisper clients, add registry, and consolidate utility functions for enhanced maintainability and extensibility.

Refactor TTS module: rename `tts_manager` to `manager` for consistency across imports and structure.

Refactor speech processing: add type hint for `_process_active_voice_frame` and replace `max()` with `np.max()` for consistency.

Refactor TTS audio generation: rename queues for clarity, update `TTSAudioGenerator` initialization, and enhance docstrings for better maintainability.

Refactor TTS architecture: implement runtime interface, TTS manager, universal registry, and factory pattern to support multiple engines.

Refactor voice model structure: extract MoYoYo-specific configurations and introduce universal TTS registry.

Update TTS inference to validate audio duration using soundfile.

Remove unused `prompt_semantic` and `reference_spec` configuration parameters from voice model definitions.

Add thread readiness checks and is_ready property across services

Update FunASR punc quantized model.

Using FunASR quantized model.

Add multilingual support and optimize LLM pipeline configuration.

Update .gitattributes to track GGUF files with git-lfs

Add descriptions for Chinese voice models.

Remove unused configuration parameters and conversation templates.

Integrate FunASR service.

Add FunASR models.

Update .gitattributes to track jieba dictionary and PNG files with git-lfs.

Update README for enhanced installation and usage guidance.

Add requirements.txt.

First commit.

Update git-lfs file track.

initial commit