SFM2001/spititout

1---2title: SPITITOUT3emoji: 🔥4colorFrom: red5colorTo: green6sdk: docker7app_port: 78608---9 10<div align="center">11<img width="1200" height="475" alt="GHBanner" src="https://github.com/user-attachments/assets/0aa67016-6eaf-458a-adb2-6e31a0763ed6" />12</div>13 14# SPITITOUT15 16This version is set up for a Hugging Face Docker Space. It does not use Gemini or any external model API; the Space loads local Hugging Face models for chat, speech recognition, and speech synthesis.17 18## Models19 20- Text on CPU Space: `Qwen/Qwen3-1.7B-GGUF` with llama.cpp / GGUF quantization21- Text on GPU Space: `Qwen/Qwen3-4B-Instruct-2507` with Transformers, or vLLM if you add a dedicated vLLM server22- Speech to text: `openai/whisper-tiny`23- Text to speech: `hexgrad/Kokoro-82M` through the `kokoro` Python package, Mandarin voice `zf_xiaobei`24 25For a CPU-only Space, keep `LLM_BACKEND=llamacpp`. To switch back to Transformers, set `LLM_BACKEND=transformers` and update `TEXT_MODEL`.26 27## Run locally28 29```bash30npm install31npm run build32pip install -r requirements.txt33python app.py34```35 36Open `http://localhost:7860`.37 38## Deploy to Hugging Face Spaces39 40Create a Docker Space, then push this folder. The included `Dockerfile` builds the React frontend and serves it from the FastAPI backend.41